| @@ -1,780 +1,779 | |||
|
|
1 | 1 | '''' |
|
|
2 | 2 | Created on Set 9, 2015 |
|
|
3 | 3 | |
|
|
4 | 4 | @author: roj-idl71 Karim Kuyeng |
|
|
5 | 5 | |
|
|
6 | 6 | @upgrade: 2021, Joab Apaza |
|
|
7 | 7 | |
|
|
8 | 8 | ''' |
|
|
9 | 9 | |
|
|
10 | 10 | import os |
|
|
11 | 11 | import sys |
|
|
12 | 12 | import glob |
|
|
13 | 13 | import fnmatch |
|
|
14 | 14 | import datetime |
|
|
15 | 15 | import time |
|
|
16 | 16 | import re |
|
|
17 | 17 | import h5py |
|
|
18 | 18 | import numpy |
|
|
19 | 19 | |
|
|
20 | 20 | try: |
|
|
21 | 21 | from gevent import sleep |
|
|
22 | 22 | except: |
|
|
23 | 23 | from time import sleep |
|
|
24 | 24 | |
|
|
25 | 25 | from schainpy.model.data.jroheaderIO import RadarControllerHeader, SystemHeader,ProcessingHeader |
|
|
26 | 26 | from schainpy.model.data.jrodata import Voltage |
|
|
27 | 27 | from schainpy.model.proc.jroproc_base import ProcessingUnit, Operation, MPDecorator |
|
|
28 | 28 | from numpy import imag |
|
|
29 | 29 | from schainpy.utils import log |
|
|
30 | 30 | |
|
|
31 | 31 | |
|
|
32 | 32 | class AMISRReader(ProcessingUnit): |
|
|
33 | 33 | ''' |
|
|
34 | 34 | classdocs |
|
|
35 | 35 | ''' |
|
|
36 | 36 | |
|
|
37 | 37 | def __init__(self): |
|
|
38 | 38 | ''' |
|
|
39 | 39 | Constructor |
|
|
40 | 40 | ''' |
|
|
41 | 41 | |
|
|
42 | 42 | ProcessingUnit.__init__(self) |
|
|
43 | 43 | |
|
|
44 | 44 | self.set = None |
|
|
45 | 45 | self.subset = None |
|
|
46 | 46 | self.extension_file = '.h5' |
|
|
47 | 47 | self.dtc_str = 'dtc' |
|
|
48 | 48 | self.dtc_id = 0 |
|
|
49 | 49 | self.status = True |
|
|
50 | 50 | self.isConfig = False |
|
|
51 | 51 | self.dirnameList = [] |
|
|
52 | 52 | self.filenameList = [] |
|
|
53 | 53 | self.fileIndex = None |
|
|
54 | 54 | self.flagNoMoreFiles = False |
|
|
55 | 55 | self.flagIsNewFile = 0 |
|
|
56 | 56 | self.filename = '' |
|
|
57 | 57 | self.amisrFilePointer = None |
|
|
58 | 58 | |
|
|
59 | 59 | self.beamCodeMap = None |
|
|
60 | 60 | self.azimuthList = [] |
|
|
61 | 61 | self.elevationList = [] |
|
|
62 | 62 | self.dataShape = None |
|
|
63 | 63 | self.flag_old_beams = False |
|
|
64 | 64 | |
|
|
65 | 65 | self.flagAsync = False #Use when the experiment has no syncronization |
|
|
66 | 66 | self.shiftChannels = 0 |
|
|
67 | 67 | self.profileIndex = 0 |
|
|
68 | 68 | |
|
|
69 | 69 | |
|
|
70 | 70 | self.beamCodeByFrame = None |
|
|
71 | 71 | self.radacTimeByFrame = None |
|
|
72 | 72 | |
|
|
73 | 73 | self.dataset = None |
|
|
74 | 74 | |
|
|
75 | 75 | self.__firstFile = True |
|
|
76 | 76 | |
|
|
77 | 77 | self.buffer = None |
|
|
78 | 78 | |
|
|
79 | 79 | self.timezone = 'ut' |
|
|
80 | 80 | |
|
|
81 | 81 | self.__waitForNewFile = 20 |
|
|
82 | 82 | self.__filename_online = None |
|
|
83 | 83 | #Is really necessary create the output object in the initializer |
|
|
84 | 84 | self.dataOut = Voltage() |
|
|
85 | 85 | self.dataOut.error=False |
|
|
86 | 86 | self.margin_days = 1 |
|
|
87 | 87 | self.flag_ignoreFiles = False #to activate the ignoring Files flag |
|
|
88 | 88 | self.flag_standby = False # just keep waiting, use when ignoring files |
|
|
89 | 89 | self.ignStartDateTime=None |
|
|
90 | 90 | self.ignEndDateTime=None |
|
|
91 | 91 | |
|
|
92 | 92 | def setup(self,path=None, |
|
|
93 | 93 | startDate=None, |
|
|
94 | 94 | endDate=None, |
|
|
95 | 95 | startTime=None, |
|
|
96 | 96 | endTime=None, |
|
|
97 | 97 | walk=True, |
|
|
98 | 98 | timezone='ut', |
|
|
99 | 99 | all=0, |
|
|
100 | 100 | code = 1, |
|
|
101 | 101 | nCode = 1, |
|
|
102 | 102 | nBaud = 0, |
|
|
103 | 103 | nOsamp = 0, |
|
|
104 | 104 | online=False, |
|
|
105 | 105 | old_beams=False, |
|
|
106 | 106 | margin_days=1, |
|
|
107 | 107 | nFFT = None, |
|
|
108 | 108 | nChannels = None, |
|
|
109 | 109 | ignStartDate=None, |
|
|
110 | 110 | ignEndDate=None, |
|
|
111 | 111 | ignStartTime=None, |
|
|
112 | 112 | ignEndTime=None, |
|
|
113 | 113 | syncronization=True, |
|
|
114 | 114 | shiftChannels=0 |
|
|
115 | 115 | ): |
|
|
116 | 116 | |
|
|
117 | 117 | |
|
|
118 | 118 | |
|
|
119 | 119 | self.timezone = timezone |
|
|
120 | 120 | self.all = all |
|
|
121 | 121 | self.online = online |
|
|
122 | 122 | self.flag_old_beams = old_beams |
|
|
123 | 123 | self.code = code |
|
|
124 | 124 | self.nCode = int(nCode) |
|
|
125 | 125 | self.nBaud = int(nBaud) |
|
|
126 | 126 | self.nOsamp = int(nOsamp) |
|
|
127 | 127 | self.margin_days = margin_days |
|
|
128 | 128 | self.__sampleRate = None |
|
|
129 | 129 | self.flagAsync = not syncronization |
|
|
130 | 130 | self.shiftChannels = shiftChannels |
|
|
131 | 131 | self.nFFT = nFFT |
|
|
132 | 132 | self.nChannels = nChannels |
|
|
133 | 133 | if ignStartTime!=None and ignEndTime!=None: |
|
|
134 | 134 | if ignStartDate!=None and ignEndDate!=None: |
|
|
135 | 135 | self.ignStartDateTime=datetime.datetime.combine(ignStartDate,ignStartTime) |
|
|
136 | 136 | self.ignEndDateTime=datetime.datetime.combine(ignEndDate,ignEndTime) |
|
|
137 | 137 | else: |
|
|
138 | 138 | self.ignStartDateTime=datetime.datetime.combine(startDate,ignStartTime) |
|
|
139 | 139 | self.ignEndDateTime=datetime.datetime.combine(endDate,ignEndTime) |
|
|
140 | 140 | self.flag_ignoreFiles = True |
|
|
141 | 141 | |
|
|
142 | 142 | #self.findFiles() |
|
|
143 | 143 | if not(online): |
|
|
144 | 144 | #Busqueda de archivos offline |
|
|
145 | 145 | self.searchFilesOffLine(path, startDate, endDate, startTime, endTime, walk,) |
|
|
146 | 146 | else: |
|
|
147 | 147 | self.searchFilesOnLine(path, startDate, endDate, startTime,endTime,walk) |
|
|
148 | 148 | |
|
|
149 | 149 | if not(self.filenameList): |
|
|
150 | 150 | raise schainpy.admin.SchainWarning("There is no files into the folder: %s"%(path)) |
|
|
151 | 151 | #sys.exit(0) |
|
|
152 | 152 | self.dataOut.error = True |
|
|
153 | 153 | |
|
|
154 | 154 | self.fileIndex = 0 |
|
|
155 | 155 | |
|
|
156 | 156 | self.readNextFile(online) |
|
|
157 | 157 | |
|
|
158 | 158 | ''' |
|
|
159 | 159 | Add code |
|
|
160 | 160 | ''' |
|
|
161 | 161 | self.isConfig = True |
|
|
162 | 162 | # print("Setup Done") |
|
|
163 | 163 | pass |
|
|
164 | 164 | |
|
|
165 | 165 | |
|
|
166 | 166 | def readAMISRHeader(self,fp): |
|
|
167 | 167 | |
|
|
168 | 168 | if self.isConfig and (not self.flagNoMoreFiles): |
|
|
169 | 169 | newShape = fp.get('Raw11/Data/Samples/Data').shape[1:] |
|
|
170 | 170 | if self.dataShape != newShape and newShape != None and not self.flag_standby: |
|
|
171 | 171 | raise schainpy.admin.SchainError("NEW FILE HAS A DIFFERENT SHAPE: ") |
|
|
172 | 172 | print(self.dataShape,newShape,"\n") |
|
|
173 | 173 | return 0 |
|
|
174 | 174 | else: |
|
|
175 | 175 | self.dataShape = fp.get('Raw11/Data/Samples/Data').shape[1:] |
|
|
176 | 176 | |
|
|
177 | 177 | |
|
|
178 | 178 | header = 'Raw11/Data/RadacHeader' |
|
|
179 | 179 | if self.nChannels == None: |
|
|
180 | 180 | expFile = fp['Setup/Experimentfile'][()].decode() |
|
|
181 | 181 | linesExp = expFile.split("\n") |
|
|
182 | 182 | a = [line for line in linesExp if "nbeamcodes" in line] |
|
|
183 | 183 | self.nChannels = int(a[0][11:]) |
|
|
184 | 184 | |
|
|
185 | 185 | if not self.flagAsync: #for experiments with no syncronization |
|
|
186 | 186 | self.shiftChannels = 0 |
|
|
187 | 187 | |
|
|
188 | 188 | |
|
|
189 | 189 | |
|
|
190 | 190 | self.beamCodeByPulse = fp.get(header+'/BeamCode') # LIST OF BEAMS PER PROFILE, TO BE USED ON REARRANGE |
|
|
191 | 191 | |
|
|
192 | 192 | |
|
|
193 | 193 | if (self.startDate > datetime.date(2021, 7, 15)) or self.flag_old_beams: #Se cambiΓ³ la forma de extracciΓ³n de Apuntes el 17 o forzar con flag de reorganizaciΓ³n |
|
|
194 | 194 | self.beamcodeFile = fp['Setup/Beamcodefile'][()].decode() |
|
|
195 | 195 | self.trueBeams = self.beamcodeFile.split("\n") |
|
|
196 | 196 | self.trueBeams.pop()#remove last |
|
|
197 | 197 | if self.nFFT == None: |
|
|
198 | 198 | log.error("FFT or number of repetitions per channels is needed",self.name) |
|
|
199 | 199 | beams_idx = [k*self.nFFT for k in range(self.nChannels)] |
|
|
200 | 200 | beams = [self.trueBeams[b] for b in beams_idx] |
|
|
201 | 201 | self.beamCode = [int(x, 16) for x in beams] |
|
|
202 | 202 | |
|
|
203 | 203 | if(self.flagAsync and self.shiftChannels == 0): |
|
|
204 | 204 | initBeam = self.beamCodeByPulse[0, 0] |
|
|
205 | 205 | self.shiftChannels = numpy.argwhere(self.beamCode ==initBeam)[0,0] |
|
|
206 | 206 | |
|
|
207 | 207 | else: |
|
|
208 | 208 | _beamCode= fp.get('Raw11/Data/Beamcodes') #se usa la manera previa al cambio de apuntes |
|
|
209 | 209 | self.beamCode = _beamCode[0,:] |
|
|
210 | 210 | |
|
|
211 | 211 | |
|
|
212 | 212 | |
|
|
213 | 213 | |
|
|
214 | 214 | if self.beamCodeMap == None: |
|
|
215 | 215 | self.beamCodeMap = fp['Setup/BeamcodeMap'] |
|
|
216 | 216 | for beam in self.beamCode: |
|
|
217 | 217 | beamAziElev = numpy.where(self.beamCodeMap[:,0]==beam) |
|
|
218 | 218 | beamAziElev = beamAziElev[0].squeeze() |
|
|
219 | 219 | self.azimuthList.append(self.beamCodeMap[beamAziElev,1]) |
|
|
220 | 220 | self.elevationList.append(self.beamCodeMap[beamAziElev,2]) |
|
|
221 | 221 | #print("Beamssss: ",self.beamCodeMap[beamAziElev,1],self.beamCodeMap[beamAziElev,2]) |
|
|
222 | 222 | #print(self.beamCode) |
|
|
223 | 223 | #self.code = fp.get(header+'/Code') # NOT USE FOR THIS |
|
|
224 | 224 | self.frameCount = fp.get(header+'/FrameCount')# NOT USE FOR THIS |
|
|
225 | 225 | self.modeGroup = fp.get(header+'/ModeGroup')# NOT USE FOR THIS |
|
|
226 | 226 | self.nsamplesPulse = fp.get(header+'/NSamplesPulse')# TO GET NSA OR USING DATA FOR THAT |
|
|
227 | 227 | self.pulseCount = fp.get(header+'/PulseCount')# NOT USE FOR THIS |
|
|
228 | 228 | self.radacTime = fp.get(header+'/RadacTime')# 1st TIME ON FILE ANDE CALCULATE THE REST WITH IPP*nindexprofile |
|
|
229 | 229 | self.timeCount = fp.get(header+'/TimeCount')# NOT USE FOR THIS |
|
|
230 | 230 | self.timeStatus = fp.get(header+'/TimeStatus')# NOT USE FOR THIS |
|
|
231 | 231 | self.rangeFromFile = fp.get('Raw11/Data/Samples/Range') |
|
|
232 | 232 | self.frequency = fp.get('Rx/Frequency') |
|
|
233 | 233 | txAus = fp.get('Raw11/Data/Pulsewidth') #seconds |
|
|
234 | 234 | self.baud = fp.get('Raw11/Data/TxBaud') |
|
|
235 | 235 | sampleRate = fp.get('Rx/SampleRate') |
|
|
236 | 236 | self.__sampleRate = sampleRate[()] |
|
|
237 | 237 | self.nblocks = self.pulseCount.shape[0] #nblocks |
|
|
238 | 238 | self.profPerBlockRAW = self.pulseCount.shape[1] #profiles per block in raw data |
|
|
239 | 239 | self.nprofiles = self.pulseCount.shape[1] #nprofile |
|
|
240 | 240 | #self.nsa = self.nsamplesPulse[0,0] #ngates |
|
|
241 | 241 | self.nsa = len(self.rangeFromFile[0]) |
|
|
242 | 242 | self.nchannels = len(self.beamCode) |
|
|
243 | 243 | self.ippSeconds = (self.radacTime[0][1] -self.radacTime[0][0]) #Ipp in seconds |
|
|
244 | 244 | #print("IPPS secs: ",self.ippSeconds) |
|
|
245 | 245 | #self.__waitForNewFile = self.nblocks # wait depending on the number of blocks since each block is 1 sec |
|
|
246 | 246 | self.__waitForNewFile = self.nblocks * self.nprofiles * self.ippSeconds # wait until new file is created |
|
|
247 | 247 | |
|
|
248 | 248 | #filling radar controller header parameters |
|
|
249 | 249 | self.__ippKm = self.ippSeconds *.15*1e6 # in km |
|
|
250 | 250 | #self.__txA = txAus[()]*.15 #(ipp[us]*.15km/1us) in km |
|
|
251 | 251 | self.__txA = txAus[()] #seconds |
|
|
252 | 252 | self.__txAKm = self.__txA*1e6*.15 |
|
|
253 | 253 | self.__txB = 0 |
|
|
254 | 254 | nWindows=1 |
|
|
255 | 255 | self.__nSamples = self.nsa |
|
|
256 | 256 | self.__firstHeight = self.rangeFromFile[0][0]/1000 #in km |
|
|
257 | 257 | self.__deltaHeight = (self.rangeFromFile[0][1] - self.rangeFromFile[0][0])/1000 |
|
|
258 | 258 | #print("amisr-ipp:",self.ippSeconds, self.__ippKm) |
|
|
259 | 259 | #for now until understand why the code saved is different (code included even though code not in tuf file) |
|
|
260 | 260 | #self.__codeType = 0 |
|
|
261 | 261 | # self.__nCode = None |
|
|
262 | 262 | # self.__nBaud = None |
|
|
263 | 263 | self.__code = self.code |
|
|
264 | 264 | self.__codeType = 0 |
|
|
265 | 265 | if self.code != None: |
|
|
266 | 266 | self.__codeType = 1 |
|
|
267 | 267 | self.__nCode = self.nCode |
|
|
268 | 268 | self.__nBaud = self.nBaud |
|
|
269 | 269 | self.__baudTX = self.__txA/(self.nBaud) |
|
|
270 | 270 | #self.__code = 0 |
|
|
271 | 271 | |
|
|
272 | 272 | #filling system header parameters |
|
|
273 | 273 | self.__nSamples = self.nsa |
|
|
274 | 274 | self.newProfiles = self.nprofiles/self.nchannels |
|
|
275 | 275 | self.__channelList = [n for n in range(self.nchannels)] |
|
|
276 | 276 | |
|
|
277 | 277 | self.__frequency = self.frequency[0][0] |
|
|
278 | 278 | |
|
|
279 | 279 | |
|
|
280 | 280 | return 1 |
|
|
281 | 281 | |
|
|
282 | 282 | |
|
|
283 | 283 | def createBuffers(self): |
|
|
284 | 284 | |
|
|
285 | 285 | pass |
|
|
286 | 286 | |
|
|
287 | 287 | def __setParameters(self,path='', startDate='',endDate='',startTime='', endTime='', walk=''): |
|
|
288 | 288 | self.path = path |
|
|
289 | 289 | self.startDate = startDate |
|
|
290 | 290 | self.endDate = endDate |
|
|
291 | 291 | self.startTime = startTime |
|
|
292 | 292 | self.endTime = endTime |
|
|
293 | 293 | self.walk = walk |
|
|
294 | 294 | |
|
|
295 | 295 | |
|
|
296 | 296 | def __checkPath(self): |
|
|
297 | 297 | if os.path.exists(self.path): |
|
|
298 | 298 | self.status = 1 |
|
|
299 | 299 | else: |
|
|
300 | 300 | self.status = 0 |
|
|
301 | 301 | print('Path:%s does not exists'%self.path) |
|
|
302 | 302 | |
|
|
303 | 303 | return |
|
|
304 | 304 | |
|
|
305 | 305 | |
|
|
306 | 306 | def __selDates(self, amisr_dirname_format): |
|
|
307 | 307 | try: |
|
|
308 | 308 | year = int(amisr_dirname_format[0:4]) |
|
|
309 | 309 | month = int(amisr_dirname_format[4:6]) |
|
|
310 | 310 | dom = int(amisr_dirname_format[6:8]) |
|
|
311 | 311 | thisDate = datetime.date(year,month,dom) |
|
|
312 | 312 | #margen de un dΓa extra, igual luego se filtra for fecha y hora |
|
|
313 | 313 | if (thisDate>=(self.startDate - datetime.timedelta(days=self.margin_days)) and thisDate <= (self.endDate)+ datetime.timedelta(days=1)): |
|
|
314 | 314 | return amisr_dirname_format |
|
|
315 | 315 | except: |
|
|
316 | 316 | return None |
|
|
317 | 317 | |
|
|
318 | 318 | |
|
|
319 | 319 | def __findDataForDates(self,online=False): |
|
|
320 | 320 | |
|
|
321 | 321 | if not(self.status): |
|
|
322 | 322 | return None |
|
|
323 | 323 | |
|
|
324 | 324 | pat = '\d+.\d+' |
|
|
325 | 325 | dirnameList = [re.search(pat,x) for x in os.listdir(self.path)] |
|
|
326 | 326 | dirnameList = [x for x in dirnameList if x!=None] |
|
|
327 | 327 | dirnameList = [x.string for x in dirnameList] |
|
|
328 | 328 | if not(online): |
|
|
329 | 329 | dirnameList = [self.__selDates(x) for x in dirnameList] |
|
|
330 | 330 | dirnameList = [x for x in dirnameList if x!=None] |
|
|
331 | 331 | if len(dirnameList)>0: |
|
|
332 | 332 | self.status = 1 |
|
|
333 | 333 | self.dirnameList = dirnameList |
|
|
334 | 334 | self.dirnameList.sort() |
|
|
335 | 335 | else: |
|
|
336 | 336 | self.status = 0 |
|
|
337 | 337 | return None |
|
|
338 | 338 | |
|
|
339 | 339 | def __getTimeFromData(self): |
|
|
340 | 340 | startDateTime_Reader = datetime.datetime.combine(self.startDate,self.startTime) |
|
|
341 | 341 | endDateTime_Reader = datetime.datetime.combine(self.endDate,self.endTime) |
|
|
342 | 342 | |
|
|
343 | 343 | print('Filtering Files from %s to %s'%(startDateTime_Reader, endDateTime_Reader)) |
|
|
344 | 344 | print('........................................') |
|
|
345 | 345 | filter_filenameList = [] |
|
|
346 | 346 | self.filenameList.sort() |
|
|
347 | 347 | total_files = len(self.filenameList) |
|
|
348 | 348 | #for i in range(len(self.filenameList)-1): |
|
|
349 | 349 | for i in range(total_files): |
|
|
350 | 350 | filename = self.filenameList[i] |
|
|
351 | 351 | #print("file-> ",filename) |
|
|
352 | 352 | try: |
|
|
353 | 353 | fp = h5py.File(filename,'r') |
|
|
354 | 354 | time_str = fp.get('Time/RadacTimeString') |
|
|
355 | 355 | |
|
|
356 | 356 | startDateTimeStr_File = time_str[0][0].decode('UTF-8').split('.')[0] |
|
|
357 | 357 | #startDateTimeStr_File = "2019-12-16 09:21:11" |
|
|
358 | 358 | junk = time.strptime(startDateTimeStr_File, '%Y-%m-%d %H:%M:%S') |
|
|
359 | 359 | startDateTime_File = datetime.datetime(junk.tm_year,junk.tm_mon,junk.tm_mday,junk.tm_hour, junk.tm_min, junk.tm_sec) |
|
|
360 | 360 | |
|
|
361 | 361 | #endDateTimeStr_File = "2019-12-16 11:10:11" |
|
|
362 | 362 | endDateTimeStr_File = time_str[-1][-1].decode('UTF-8').split('.')[0] |
|
|
363 | 363 | junk = time.strptime(endDateTimeStr_File, '%Y-%m-%d %H:%M:%S') |
|
|
364 | 364 | endDateTime_File = datetime.datetime(junk.tm_year,junk.tm_mon,junk.tm_mday,junk.tm_hour, junk.tm_min, junk.tm_sec) |
|
|
365 | 365 | |
|
|
366 | 366 | fp.close() |
|
|
367 | 367 | |
|
|
368 | 368 | #print("check time", startDateTime_File) |
|
|
369 | 369 | if self.timezone == 'lt': |
|
|
370 | 370 | startDateTime_File = startDateTime_File - datetime.timedelta(minutes = 300) |
|
|
371 | 371 | endDateTime_File = endDateTime_File - datetime.timedelta(minutes = 300) |
|
|
372 | 372 | if (startDateTime_File >=startDateTime_Reader and endDateTime_File<=endDateTime_Reader): |
|
|
373 | 373 | filter_filenameList.append(filename) |
|
|
374 | 374 | |
|
|
375 | 375 | if (startDateTime_File>endDateTime_Reader): |
|
|
376 | 376 | break |
|
|
377 | 377 | except Exception as e: |
|
|
378 | 378 | log.warning("Error opening file {} -> {}".format(os.path.split(filename)[1],e)) |
|
|
379 | 379 | |
|
|
380 | 380 | filter_filenameList.sort() |
|
|
381 | 381 | self.filenameList = filter_filenameList |
|
|
382 | 382 | |
|
|
383 | 383 | return 1 |
|
|
384 | 384 | |
|
|
385 | 385 | def __filterByGlob1(self, dirName): |
|
|
386 | 386 | filter_files = glob.glob1(dirName, '*.*%s'%self.extension_file) |
|
|
387 | 387 | filter_files.sort() |
|
|
388 | 388 | filterDict = {} |
|
|
389 | 389 | filterDict.setdefault(dirName) |
|
|
390 | 390 | filterDict[dirName] = filter_files |
|
|
391 | 391 | return filterDict |
|
|
392 | 392 | |
|
|
393 | 393 | def __getFilenameList(self, fileListInKeys, dirList): |
|
|
394 | 394 | for value in fileListInKeys: |
|
|
395 | 395 | dirName = list(value.keys())[0] |
|
|
396 | 396 | for file in value[dirName]: |
|
|
397 | 397 | filename = os.path.join(dirName, file) |
|
|
398 | 398 | self.filenameList.append(filename) |
|
|
399 | 399 | |
|
|
400 | 400 | |
|
|
401 | 401 | def __selectDataForTimes(self, online=False): |
|
|
402 | 402 | #aun no esta implementado el filtro for tiempo-> implementado en readNextFile |
|
|
403 | 403 | if not(self.status): |
|
|
404 | 404 | return None |
|
|
405 | 405 | |
|
|
406 | 406 | dirList = [os.path.join(self.path,x) for x in self.dirnameList] |
|
|
407 | 407 | fileListInKeys = [self.__filterByGlob1(x) for x in dirList] |
|
|
408 | 408 | self.__getFilenameList(fileListInKeys, dirList) |
|
|
409 | 409 | if not(online): |
|
|
410 | 410 | #filtro por tiempo |
|
|
411 | 411 | if not(self.all): |
|
|
412 | 412 | self.__getTimeFromData() |
|
|
413 | 413 | |
|
|
414 | 414 | if len(self.filenameList)>0: |
|
|
415 | 415 | self.status = 1 |
|
|
416 | 416 | self.filenameList.sort() |
|
|
417 | 417 | else: |
|
|
418 | 418 | self.status = 0 |
|
|
419 | 419 | return None |
|
|
420 | 420 | |
|
|
421 | 421 | else: |
|
|
422 | 422 | #get the last file - 1 |
|
|
423 | 423 | self.filenameList = [self.filenameList[-2]] |
|
|
424 | 424 | new_dirnameList = [] |
|
|
425 | 425 | for dirname in self.dirnameList: |
|
|
426 | 426 | junk = numpy.array([dirname in x for x in self.filenameList]) |
|
|
427 | 427 | junk_sum = junk.sum() |
|
|
428 | 428 | if junk_sum > 0: |
|
|
429 | 429 | new_dirnameList.append(dirname) |
|
|
430 | 430 | self.dirnameList = new_dirnameList |
|
|
431 | 431 | return 1 |
|
|
432 | 432 | |
|
|
433 | 433 | def searchFilesOnLine(self, path, startDate, endDate, startTime=datetime.time(0,0,0), |
|
|
434 | 434 | endTime=datetime.time(23,59,59),walk=True): |
|
|
435 | 435 | |
|
|
436 | 436 | if endDate ==None: |
|
|
437 | 437 | startDate = datetime.datetime.utcnow().date() |
|
|
438 | 438 | endDate = datetime.datetime.utcnow().date() |
|
|
439 | 439 | |
|
|
440 | 440 | self.__setParameters(path=path, startDate=startDate, endDate=endDate,startTime = startTime,endTime=endTime, walk=walk) |
|
|
441 | 441 | |
|
|
442 | 442 | self.__checkPath() |
|
|
443 | 443 | |
|
|
444 | 444 | self.__findDataForDates(online=True) |
|
|
445 | 445 | |
|
|
446 | 446 | self.dirnameList = [self.dirnameList[-1]] |
|
|
447 | 447 | |
|
|
448 | 448 | self.__selectDataForTimes(online=True) |
|
|
449 | 449 | |
|
|
450 | 450 | return |
|
|
451 | 451 | |
|
|
452 | 452 | |
|
|
453 | 453 | def searchFilesOffLine(self, |
|
|
454 | 454 | path, |
|
|
455 | 455 | startDate, |
|
|
456 | 456 | endDate, |
|
|
457 | 457 | startTime=datetime.time(0,0,0), |
|
|
458 | 458 | endTime=datetime.time(23,59,59), |
|
|
459 | 459 | walk=True): |
|
|
460 | 460 | |
|
|
461 | 461 | self.__setParameters(path, startDate, endDate, startTime, endTime, walk) |
|
|
462 | 462 | |
|
|
463 | 463 | self.__checkPath() |
|
|
464 | 464 | |
|
|
465 | 465 | self.__findDataForDates() |
|
|
466 | 466 | |
|
|
467 | 467 | self.__selectDataForTimes() |
|
|
468 | 468 | |
|
|
469 | 469 | for i in range(len(self.filenameList)): |
|
|
470 | 470 | print("%s" %(self.filenameList[i])) |
|
|
471 | 471 | |
|
|
472 | 472 | return |
|
|
473 | 473 | |
|
|
474 | 474 | def __setNextFileOffline(self): |
|
|
475 | 475 | |
|
|
476 | 476 | try: |
|
|
477 | 477 | self.filename = self.filenameList[self.fileIndex] |
|
|
478 | 478 | self.amisrFilePointer = h5py.File(self.filename,'r') |
|
|
479 | 479 | self.fileIndex += 1 |
|
|
480 | 480 | except: |
|
|
481 | 481 | self.flagNoMoreFiles = 1 |
|
|
482 | 482 | raise schainpy.admin.SchainError('No more files to read') |
|
|
483 | 483 | return 0 |
|
|
484 | 484 | |
|
|
485 | 485 | self.flagIsNewFile = 1 |
|
|
486 | 486 | print("Setting the file: %s"%self.filename) |
|
|
487 | 487 | |
|
|
488 | 488 | return 1 |
|
|
489 | 489 | |
|
|
490 | 490 | |
|
|
491 | 491 | def __setNextFileOnline(self): |
|
|
492 | 492 | filename = self.filenameList[0] |
|
|
493 | 493 | if self.__filename_online != None: |
|
|
494 | 494 | self.__selectDataForTimes(online=True) |
|
|
495 | 495 | filename = self.filenameList[0] |
|
|
496 | 496 | wait = 0 |
|
|
497 | 497 | self.__waitForNewFile=300 ## DEBUG: |
|
|
498 | 498 | while self.__filename_online == filename: |
|
|
499 | 499 | print('waiting %d seconds to get a new file...'%(self.__waitForNewFile)) |
|
|
500 | 500 | if wait == 5: |
|
|
501 | 501 | self.flagNoMoreFiles = 1 |
|
|
502 | 502 | return 0 |
|
|
503 | 503 | sleep(self.__waitForNewFile) |
|
|
504 | 504 | self.__selectDataForTimes(online=True) |
|
|
505 | 505 | filename = self.filenameList[0] |
|
|
506 | 506 | wait += 1 |
|
|
507 | 507 | |
|
|
508 | 508 | self.__filename_online = filename |
|
|
509 | 509 | |
|
|
510 | 510 | self.amisrFilePointer = h5py.File(filename,'r') |
|
|
511 | 511 | self.flagIsNewFile = 1 |
|
|
512 | 512 | self.filename = filename |
|
|
513 | 513 | print("Setting the file: %s"%self.filename) |
|
|
514 | 514 | return 1 |
|
|
515 | 515 | |
|
|
516 | 516 | |
|
|
517 | 517 | def readData(self): |
|
|
518 | 518 | buffer = self.amisrFilePointer.get('Raw11/Data/Samples/Data') |
|
|
519 | 519 | re = buffer[:,:,:,0] |
|
|
520 | 520 | im = buffer[:,:,:,1] |
|
|
521 | 521 | dataset = re + im*1j |
|
|
522 | 522 | |
|
|
523 | 523 | self.radacTime = self.amisrFilePointer.get('Raw11/Data/RadacHeader/RadacTime') |
|
|
524 | 524 | timeset = self.radacTime[:,0] |
|
|
525 | 525 | |
|
|
526 | 526 | return dataset,timeset |
|
|
527 | 527 | |
|
|
528 | 528 | def reshapeData(self): |
|
|
529 | 529 | #print(self.beamCodeByPulse, self.beamCode, self.nblocks, self.nprofiles, self.nsa) |
|
|
530 | 530 | channels = self.beamCodeByPulse[0,:] |
|
|
531 | 531 | nchan = self.nchannels |
|
|
532 | 532 | #self.newProfiles = self.nprofiles/nchan #must be defined on filljroheader |
|
|
533 | 533 | nblocks = self.nblocks |
|
|
534 | 534 | nsamples = self.nsa |
|
|
535 | 535 | #print("Channels: ",self.nChannels) |
|
|
536 | print("dataset: ", self.dataset.shape) | |
|
|
537 | 536 | #Dimensions : nChannels, nProfiles, nSamples |
|
|
538 | 537 | new_block = numpy.empty((nblocks, nchan, numpy.int_(self.newProfiles), nsamples), dtype="complex64") |
|
|
539 | 538 | ############################################ |
|
|
540 | 539 | profPerCH = int(self.profPerBlockRAW / (self.nFFT* self.nChannels)) |
|
|
541 | 540 | #profPerCH = int(self.profPerBlockRAW / self.nChannels) |
|
|
542 | 541 | for thisChannel in range(nchan): |
|
|
543 | 542 | |
|
|
544 | 543 | ich = thisChannel |
|
|
545 | 544 | |
|
|
546 | 545 | idx_ch = [self.nFFT*(ich + nchan*k) for k in range(profPerCH)] |
|
|
547 | 546 | #print(idx_ch) |
|
|
548 | 547 | if self.nFFT > 1: |
|
|
549 | 548 | aux = [numpy.arange(i, i+self.nFFT) for i in idx_ch] |
|
|
550 | 549 | idx_ch = None |
|
|
551 | 550 | idx_ch =aux |
|
|
552 | 551 | idx_ch = numpy.array(idx_ch, dtype=int).flatten() |
|
|
553 | 552 | else: |
|
|
554 | 553 | idx_ch = numpy.array(idx_ch, dtype=int) |
|
|
555 | 554 | |
|
|
556 | 555 | #print(ich,profPerCH,idx_ch) |
|
|
557 | 556 | #print(numpy.where(channels==self.beamCode[ich])[0]) |
|
|
558 | 557 | #new_block[:,ich,:,:] = self.dataset[:,numpy.where(channels==self.beamCode[ich])[0],:] |
|
|
559 | 558 | new_block[:,ich,:,:] = self.dataset[:,idx_ch,:] |
|
|
560 | 559 | |
|
|
561 | 560 | new_block = numpy.transpose(new_block, (1,0,2,3)) |
|
|
562 | 561 | new_block = numpy.reshape(new_block, (nchan,-1, nsamples)) |
|
|
563 | 562 | if self.flagAsync: |
|
|
564 | 563 | new_block = numpy.roll(new_block, self.shiftChannels, axis=0) |
|
|
565 | 564 | return new_block |
|
|
566 | 565 | |
|
|
567 | 566 | def updateIndexes(self): |
|
|
568 | 567 | |
|
|
569 | 568 | pass |
|
|
570 | 569 | |
|
|
571 | 570 | def fillJROHeader(self): |
|
|
572 | 571 | |
|
|
573 | 572 | #fill radar controller header |
|
|
574 | 573 | |
|
|
575 | 574 | #fill system header |
|
|
576 | 575 | self.dataOut.systemHeaderObj = SystemHeader(nSamples=self.__nSamples, |
|
|
577 | 576 | nProfiles=self.newProfiles, |
|
|
578 | 577 | nChannels=len(self.__channelList), |
|
|
579 | 578 | adcResolution=14, |
|
|
580 | 579 | pciDioBusWidth=32) |
|
|
581 | 580 | |
|
|
582 | 581 | self.dataOut.type = "Voltage" |
|
|
583 | 582 | self.dataOut.data = None |
|
|
584 | 583 | self.dataOut.dtype = numpy.dtype([('real','<i8'),('imag','<i8')]) |
|
|
585 | 584 | # self.dataOut.nChannels = 0 |
|
|
586 | 585 | |
|
|
587 | 586 | # self.dataOut.nHeights = 0 |
|
|
588 | 587 | |
|
|
589 | 588 | self.dataOut.nProfiles = self.newProfiles*self.nblocks |
|
|
590 | 589 | #self.dataOut.heightList = self.__firstHeigth + numpy.arange(self.__nSamples, dtype = numpy.float)*self.__deltaHeigth |
|
|
591 | 590 | ranges = numpy.reshape(self.rangeFromFile[()],(-1)) |
|
|
592 | 591 | self.dataOut.heightList = ranges/1000.0 #km |
|
|
593 | 592 | self.dataOut.channelList = self.__channelList |
|
|
594 | 593 | |
|
|
595 | 594 | self.dataOut.blocksize = self.dataOut.nChannels * self.dataOut.nHeights |
|
|
596 | 595 | |
|
|
597 | 596 | # self.dataOut.channelIndexList = None |
|
|
598 | 597 | |
|
|
599 | 598 | |
|
|
600 | 599 | # #self.dataOut.azimuthList = numpy.roll( numpy.array(self.azimuthList) ,self.shiftChannels) |
|
|
601 | 600 | # #self.dataOut.elevationList = numpy.roll(numpy.array(self.elevationList) ,self.shiftChannels) |
|
|
602 | 601 | # #self.dataOut.codeList = numpy.roll(numpy.array(self.beamCode), self.shiftChannels) |
|
|
603 | 602 | |
|
|
604 | 603 | self.dataOut.azimuthList = self.azimuthList |
|
|
605 | 604 | self.dataOut.elevationList = self.elevationList |
|
|
606 | 605 | self.dataOut.codeList = self.beamCode |
|
|
607 | 606 | |
|
|
608 | 607 | |
|
|
609 | 608 | |
|
|
610 | 609 | #print(self.dataOut.elevationList) |
|
|
611 | 610 | self.dataOut.flagNoData = True |
|
|
612 | 611 | |
|
|
613 | 612 | #Set to TRUE if the data is discontinuous |
|
|
614 | 613 | self.dataOut.flagDiscontinuousBlock = False |
|
|
615 | 614 | |
|
|
616 | 615 | self.dataOut.utctime = None |
|
|
617 | 616 | |
|
|
618 | 617 | #self.dataOut.timeZone = -5 #self.__timezone/60 #timezone like jroheader, difference in minutes between UTC and localtime |
|
|
619 | 618 | if self.timezone == 'lt': |
|
|
620 | 619 | self.dataOut.timeZone = time.timezone / 60. #get the timezone in minutes |
|
|
621 | 620 | else: |
|
|
622 | 621 | self.dataOut.timeZone = 0 #by default time is UTC |
|
|
623 | 622 | |
|
|
624 | 623 | self.dataOut.dstFlag = 0 |
|
|
625 | 624 | self.dataOut.errorCount = 0 |
|
|
626 | 625 | self.dataOut.nCohInt = 1 |
|
|
627 | 626 | self.dataOut.flagDecodeData = False #asumo que la data esta decodificada |
|
|
628 | 627 | self.dataOut.flagDeflipData = False #asumo que la data esta sin flip |
|
|
629 | 628 | self.dataOut.flagShiftFFT = False |
|
|
630 | 629 | self.dataOut.ippSeconds = self.ippSeconds |
|
|
631 | 630 | self.dataOut.ipp = self.__ippKm |
|
|
632 | 631 | self.dataOut.nCode = self.__nCode |
|
|
633 | 632 | self.dataOut.code = self.__code |
|
|
634 | 633 | self.dataOut.nBaud = self.__nBaud |
|
|
635 | 634 | |
|
|
636 | 635 | |
|
|
637 | 636 | self.dataOut.frequency = self.__frequency |
|
|
638 | 637 | self.dataOut.realtime = self.online |
|
|
639 | 638 | |
|
|
640 | 639 | self.dataOut.radarControllerHeaderObj = RadarControllerHeader(ipp=self.__ippKm, |
|
|
641 | 640 | txA=self.__txAKm, |
|
|
642 | 641 | txB=0, |
|
|
643 | 642 | nWindows=1, |
|
|
644 | 643 | nHeights=self.__nSamples, |
|
|
645 | 644 | firstHeight=self.__firstHeight, |
|
|
646 | 645 | codeType=self.__codeType, |
|
|
647 | 646 | nCode=self.__nCode, nBaud=self.__nBaud, |
|
|
648 | 647 | code = self.__code, |
|
|
649 | 648 | nOsamp=self.nOsamp, |
|
|
650 | 649 | frequency = self.__frequency, |
|
|
651 | 650 | sampleRate= self.__sampleRate, |
|
|
652 | 651 | fClock=self.__sampleRate) |
|
|
653 | 652 | |
|
|
654 | 653 | |
|
|
655 | 654 | self.dataOut.radarControllerHeaderObj.heightList = ranges/1000.0 #km |
|
|
656 | 655 | self.dataOut.radarControllerHeaderObj.heightResolution = self.__deltaHeight |
|
|
657 | 656 | self.dataOut.radarControllerHeaderObj.rangeIpp = self.__ippKm #km |
|
|
658 | 657 | self.dataOut.radarControllerHeaderObj.rangeTxA = self.__txA*1e6*.15 #km |
|
|
659 | 658 | self.dataOut.radarControllerHeaderObj.nChannels = self.nchannels |
|
|
660 | 659 | self.dataOut.radarControllerHeaderObj.channelList = self.__channelList |
|
|
661 | 660 | self.dataOut.radarControllerHeaderObj.azimuthList = self.azimuthList |
|
|
662 | 661 | self.dataOut.radarControllerHeaderObj.elevationList = self.elevationList |
|
|
663 | 662 | self.dataOut.radarControllerHeaderObj.dtype = "Voltage" |
|
|
664 | 663 | self.dataOut.ippSeconds = self.ippSeconds |
|
|
665 | 664 | self.dataOut.ippFactor = self.nFFT |
|
|
666 | 665 | pass |
|
|
667 | 666 | |
|
|
668 | 667 | def readNextFile(self,online=False): |
|
|
669 | 668 | |
|
|
670 | 669 | if not(online): |
|
|
671 | 670 | newFile = self.__setNextFileOffline() |
|
|
672 | 671 | else: |
|
|
673 | 672 | newFile = self.__setNextFileOnline() |
|
|
674 | 673 | |
|
|
675 | 674 | if not(newFile): |
|
|
676 | 675 | self.dataOut.error = True |
|
|
677 | 676 | return 0 |
|
|
678 | 677 | |
|
|
679 | 678 | if not self.readAMISRHeader(self.amisrFilePointer): |
|
|
680 | 679 | self.dataOut.error = True |
|
|
681 | 680 | return 0 |
|
|
682 | 681 | |
|
|
683 | 682 | #self.createBuffers() |
|
|
684 | 683 | self.fillJROHeader() |
|
|
685 | 684 | |
|
|
686 | 685 | #self.__firstFile = False |
|
|
687 | 686 | |
|
|
688 | 687 | self.dataset,self.timeset = self.readData() |
|
|
689 | 688 | |
|
|
690 | 689 | if self.endDate!=None: |
|
|
691 | 690 | endDateTime_Reader = datetime.datetime.combine(self.endDate,self.endTime) |
|
|
692 | 691 | time_str = self.amisrFilePointer.get('Time/RadacTimeString') |
|
|
693 | 692 | startDateTimeStr_File = time_str[0][0].decode('UTF-8').split('.')[0] |
|
|
694 | 693 | junk = time.strptime(startDateTimeStr_File, '%Y-%m-%d %H:%M:%S') |
|
|
695 | 694 | startDateTime_File = datetime.datetime(junk.tm_year,junk.tm_mon,junk.tm_mday,junk.tm_hour, junk.tm_min, junk.tm_sec) |
|
|
696 | 695 | if self.timezone == 'lt': |
|
|
697 | 696 | startDateTime_File = startDateTime_File - datetime.timedelta(minutes = 300) |
|
|
698 | 697 | if (startDateTime_File>endDateTime_Reader): |
|
|
699 | 698 | self.flag_standby = False |
|
|
700 | 699 | return 0 |
|
|
701 | 700 | if self.flag_ignoreFiles and (startDateTime_File >= self.ignStartDateTime and startDateTime_File <= self.ignEndDateTime): |
|
|
702 | 701 | print("Ignoring...") |
|
|
703 | 702 | self.flag_standby = True |
|
|
704 | 703 | return 1 |
|
|
705 | 704 | self.flag_standby = False |
|
|
706 | 705 | |
|
|
707 | 706 | self.jrodataset = self.reshapeData() |
|
|
708 | 707 | #----self.updateIndexes() |
|
|
709 | 708 | self.profileIndex = 0 |
|
|
710 | 709 | |
|
|
711 | 710 | return 1 |
|
|
712 | 711 | |
|
|
713 | 712 | |
|
|
714 | 713 | def __hasNotDataInBuffer(self): |
|
|
715 | 714 | if self.profileIndex >= (self.newProfiles*self.nblocks): |
|
|
716 | 715 | return 1 |
|
|
717 | 716 | return 0 |
|
|
718 | 717 | |
|
|
719 | 718 | |
|
|
720 | 719 | def getData(self): |
|
|
721 | 720 | |
|
|
722 | 721 | if self.flagNoMoreFiles: |
|
|
723 | 722 | self.dataOut.flagNoData = True |
|
|
724 | 723 | return 0 |
|
|
725 | 724 | |
|
|
726 | 725 | if self.profileIndex >= (self.newProfiles*self.nblocks): # |
|
|
727 | 726 | #if self.__hasNotDataInBuffer(): |
|
|
728 | 727 | if not (self.readNextFile(self.online)): |
|
|
729 | 728 | print("Profile Index break...") |
|
|
730 | 729 | return 0 |
|
|
731 | 730 | |
|
|
732 | 731 | if self.flag_standby: #Standby mode, if files are being ignoring, just return with no error flag |
|
|
733 | 732 | return 0 |
|
|
734 | 733 | |
|
|
735 | 734 | if self.dataset is None: # setear esta condicion cuando no hayan datos por leer |
|
|
736 | 735 | self.dataOut.flagNoData = True |
|
|
737 | 736 | print("No more data break...") |
|
|
738 | 737 | return 0 |
|
|
739 | 738 | |
|
|
740 | 739 | #self.dataOut.data = numpy.reshape(self.jrodataset[self.profileIndex,:],(1,-1)) |
|
|
741 | 740 | |
|
|
742 | 741 | self.dataOut.data = self.jrodataset[:,self.profileIndex,:] |
|
|
743 | 742 | |
|
|
744 | 743 | #print("R_t",self.timeset) |
|
|
745 | 744 | |
|
|
746 | 745 | #self.dataOut.utctime = self.jrotimeset[self.profileIndex] |
|
|
747 | 746 | #verificar basic header de jro data y ver si es compatible con este valor |
|
|
748 | 747 | #self.dataOut.utctime = self.timeset + (self.profileIndex * self.ippSeconds * self.nchannels) |
|
|
749 | 748 | indexprof = numpy.mod(self.profileIndex, self.newProfiles) |
|
|
750 | 749 | indexblock = self.profileIndex/self.newProfiles |
|
|
751 | 750 | #print (indexblock, indexprof) |
|
|
752 | 751 | diffUTC = 0 |
|
|
753 | 752 | t_comp = (indexprof * self.ippSeconds * self.nchannels) + diffUTC # |
|
|
754 | 753 | |
|
|
755 | 754 | #print("utc :",indexblock," __ ",t_comp) |
|
|
756 | 755 | #print(numpy.shape(self.timeset)) |
|
|
757 | 756 | self.dataOut.utctime = self.timeset[numpy.int_(indexblock)] + t_comp |
|
|
758 | 757 | #self.dataOut.utctime = self.timeset[self.profileIndex] + t_comp |
|
|
759 | 758 | |
|
|
760 | 759 | self.dataOut.profileIndex = self.profileIndex |
|
|
761 | 760 | #print("N profile:",self.profileIndex,self.newProfiles,self.nblocks,self.dataOut.utctime) |
|
|
762 | 761 | self.dataOut.flagNoData = False |
|
|
763 | 762 | # if indexprof == 0: |
|
|
764 | 763 | # print("kamisr: ",self.dataOut.utctime) |
|
|
765 | 764 | |
|
|
766 | 765 | self.profileIndex += 1 |
|
|
767 | 766 | |
|
|
768 | 767 | return self.dataOut.data #retorno necesario?? |
|
|
769 | 768 | |
|
|
770 | 769 | |
|
|
771 | 770 | def run(self, **kwargs): |
|
|
772 | 771 | ''' |
|
|
773 | 772 | This method will be called many times so here you should put all your code |
|
|
774 | 773 | ''' |
|
|
775 | 774 | #print("running kamisr") |
|
|
776 | 775 | if not self.isConfig: |
|
|
777 | 776 | self.setup(**kwargs) |
|
|
778 | 777 | self.isConfig = True |
|
|
779 | 778 | |
|
|
780 | 779 | self.getData() |
| @@ -1,2306 +1,2305 | |||
|
|
1 | 1 | # Copyright (c) 2012-2020 Jicamarca Radio Observatory |
|
|
2 | 2 | # All rights reserved. |
|
|
3 | 3 | # |
|
|
4 | 4 | # Distributed under the terms of the BSD 3-clause license. |
|
|
5 | 5 | """Spectra processing Unit and operations |
|
|
6 | 6 | |
|
|
7 | 7 | Here you will find the processing unit `SpectraProc` and several operations |
|
|
8 | 8 | to work with Spectra data type |
|
|
9 | 9 | """ |
|
|
10 | 10 | |
|
|
11 | 11 | import time |
|
|
12 | 12 | import itertools |
|
|
13 | 13 | |
|
|
14 | 14 | import numpy |
|
|
15 | 15 | import math |
|
|
16 | 16 | |
|
|
17 | 17 | from schainpy.model.proc.jroproc_base import ProcessingUnit, MPDecorator, Operation |
|
|
18 | 18 | from schainpy.model.data.jrodata import Spectra |
|
|
19 | 19 | from schainpy.model.data.jrodata import hildebrand_sekhon |
|
|
20 | 20 | from schainpy.model.data import _noise |
|
|
21 | 21 | |
|
|
22 | 22 | from schainpy.utils import log |
|
|
23 | 23 | import matplotlib.pyplot as plt |
|
|
24 | 24 | #from scipy.optimize import curve_fit |
|
|
25 | 25 | from schainpy.model.io.utilsIO import getHei_index |
|
|
26 | 26 | import datetime |
|
|
27 | 27 | |
|
|
28 | 28 | class SpectraProc(ProcessingUnit): |
|
|
29 | 29 | |
|
|
30 | 30 | def __init__(self): |
|
|
31 | 31 | |
|
|
32 | 32 | ProcessingUnit.__init__(self) |
|
|
33 | 33 | |
|
|
34 | 34 | self.buffer = None |
|
|
35 | 35 | self.firstdatatime = None |
|
|
36 | 36 | self.profIndex = 0 |
|
|
37 | 37 | self.dataOut = Spectra() |
|
|
38 | 38 | self.id_min = None |
|
|
39 | 39 | self.id_max = None |
|
|
40 | 40 | self.setupReq = False #Agregar a todas las unidades de proc |
|
|
41 | 41 | self.nsamplesFFT = 0 |
|
|
42 | 42 | |
|
|
43 | 43 | def __updateSpecFromVoltage(self): |
|
|
44 | 44 | |
|
|
45 | 45 | |
|
|
46 | 46 | |
|
|
47 | 47 | self.dataOut.timeZone = self.dataIn.timeZone |
|
|
48 | 48 | self.dataOut.dstFlag = self.dataIn.dstFlag |
|
|
49 | 49 | self.dataOut.errorCount = self.dataIn.errorCount |
|
|
50 | 50 | self.dataOut.useLocalTime = self.dataIn.useLocalTime |
|
|
51 | 51 | |
|
|
52 | 52 | self.dataOut.processingHeaderObj = self.dataIn.processingHeaderObj.copy() |
|
|
53 | 53 | self.dataOut.radarControllerHeaderObj = self.dataIn.radarControllerHeaderObj.copy() |
|
|
54 | 54 | self.dataOut.ippSeconds = self.dataIn.ippSeconds |
|
|
55 | 55 | self.dataOut.ipp = self.dataIn.ipp |
|
|
56 | 56 | self.dataOut.systemHeaderObj = self.dataIn.systemHeaderObj.copy() |
|
|
57 | 57 | self.dataOut.channelList = self.dataIn.channelList |
|
|
58 | 58 | self.dataOut.heightList = self.dataIn.heightList |
|
|
59 | 59 | self.dataOut.dtype = numpy.dtype([('real', '<f4'), ('imag', '<f4')]) |
|
|
60 | 60 | self.dataOut.nProfiles = self.dataOut.nFFTPoints |
|
|
61 | 61 | self.dataOut.flagDiscontinuousBlock = self.dataIn.flagDiscontinuousBlock |
|
|
62 | 62 | self.dataOut.utctime = self.firstdatatime |
|
|
63 | 63 | self.dataOut.flagDecodeData = self.dataIn.flagDecodeData |
|
|
64 | 64 | self.dataOut.flagDeflipData = self.dataIn.flagDeflipData |
|
|
65 | 65 | self.dataOut.flagShiftFFT = False |
|
|
66 | 66 | self.dataOut.nCohInt = self.dataIn.nCohInt |
|
|
67 | 67 | self.dataOut.nIncohInt = 1 |
|
|
68 | 68 | self.dataOut.deltaHeight = self.dataIn.deltaHeight |
|
|
69 | 69 | self.dataOut.windowOfFilter = self.dataIn.windowOfFilter |
|
|
70 | 70 | self.dataOut.frequency = self.dataIn.frequency |
|
|
71 | 71 | self.dataOut.realtime = self.dataIn.realtime |
|
|
72 | 72 | self.dataOut.azimuth = self.dataIn.azimuth |
|
|
73 | 73 | self.dataOut.zenith = self.dataIn.zenith |
|
|
74 | 74 | self.dataOut.codeList = self.dataIn.codeList |
|
|
75 | 75 | self.dataOut.azimuthList = self.dataIn.azimuthList |
|
|
76 | 76 | self.dataOut.elevationList = self.dataIn.elevationList |
|
|
77 | 77 | self.dataOut.code = self.dataIn.code |
|
|
78 | 78 | self.dataOut.nCode = self.dataIn.nCode |
|
|
79 | 79 | self.dataOut.flagProfilesByRange = self.dataIn.flagProfilesByRange |
|
|
80 | 80 | self.dataOut.nProfilesByRange = self.dataIn.nProfilesByRange |
|
|
81 | 81 | |
|
|
82 | 82 | |
|
|
83 | 83 | def __getFft(self): |
|
|
84 | 84 | # print("fft donw") |
|
|
85 | 85 | """ |
|
|
86 | 86 | Convierte valores de Voltaje a Spectra |
|
|
87 | 87 | |
|
|
88 | 88 | Affected: |
|
|
89 | 89 | self.dataOut.data_spc |
|
|
90 | 90 | self.dataOut.data_cspc |
|
|
91 | 91 | self.dataOut.data_dc |
|
|
92 | 92 | self.dataOut.heightList |
|
|
93 | 93 | self.profIndex |
|
|
94 | 94 | self.buffer |
|
|
95 | 95 | self.dataOut.flagNoData |
|
|
96 | 96 | """ |
|
|
97 | 97 | fft_volt = numpy.fft.fft( |
|
|
98 | 98 | self.buffer, n=self.dataOut.nFFTPoints, axis=1) |
|
|
99 | 99 | fft_volt = fft_volt.astype(numpy.dtype('complex')) |
|
|
100 | 100 | dc = fft_volt[:, 0, :] |
|
|
101 | 101 | |
|
|
102 | 102 | # calculo de self-spectra |
|
|
103 | 103 | fft_volt = numpy.fft.fftshift(fft_volt, axes=(1,)) |
|
|
104 | 104 | spc = fft_volt * numpy.conjugate(fft_volt) |
|
|
105 | 105 | spc = spc.real |
|
|
106 | 106 | |
|
|
107 | 107 | blocksize = 0 |
|
|
108 | 108 | blocksize += dc.size |
|
|
109 | 109 | blocksize += spc.size |
|
|
110 | 110 | |
|
|
111 | 111 | cspc = None |
|
|
112 | 112 | pairIndex = 0 |
|
|
113 | 113 | if self.dataOut.pairsList != None: |
|
|
114 | 114 | # calculo de cross-spectra |
|
|
115 | 115 | cspc = numpy.zeros( |
|
|
116 | 116 | (self.dataOut.nPairs, self.dataOut.nFFTPoints, self.dataOut.nHeights), dtype='complex') |
|
|
117 | 117 | for pair in self.dataOut.pairsList: |
|
|
118 | 118 | if pair[0] not in self.dataOut.channelList: |
|
|
119 | 119 | raise ValueError("Error getting CrossSpectra: pair 0 of %s is not in channelList = %s" % ( |
|
|
120 | 120 | str(pair), str(self.dataOut.channelList))) |
|
|
121 | 121 | if pair[1] not in self.dataOut.channelList: |
|
|
122 | 122 | raise ValueError("Error getting CrossSpectra: pair 1 of %s is not in channelList = %s" % ( |
|
|
123 | 123 | str(pair), str(self.dataOut.channelList))) |
|
|
124 | 124 | |
|
|
125 | 125 | cspc[pairIndex, :, :] = fft_volt[pair[0], :, :] * \ |
|
|
126 | 126 | numpy.conjugate(fft_volt[pair[1], :, :]) |
|
|
127 | 127 | pairIndex += 1 |
|
|
128 | 128 | blocksize += cspc.size |
|
|
129 | 129 | |
|
|
130 | 130 | self.dataOut.data_spc = spc |
|
|
131 | 131 | self.dataOut.data_cspc = cspc |
|
|
132 | 132 | self.dataOut.data_dc = dc |
|
|
133 | 133 | self.dataOut.blockSize = blocksize |
|
|
134 | 134 | self.dataOut.flagShiftFFT = False |
|
|
135 | 135 | |
|
|
136 | 136 | def run(self, nProfiles=None, nFFTPoints=None, pairsList=None, ippFactor=None, shift_fft=False, zeroPad=False, zeroPoints=0): |
|
|
137 | 137 | |
|
|
138 | 138 | |
|
|
139 | 139 | try: |
|
|
140 | 140 | type = self.dataIn.type.decode("utf-8") |
|
|
141 | 141 | self.dataIn.type = type |
|
|
142 | 142 | except: |
|
|
143 | 143 | pass |
|
|
144 | 144 | if self.dataIn.type == "Spectra": |
|
|
145 | 145 | |
|
|
146 | 146 | try: |
|
|
147 | 147 | self.dataOut.copy(self.dataIn) |
|
|
148 | 148 | self.dataOut.radarControllerHeaderObj = self.dataIn.radarControllerHeaderObj.copy() |
|
|
149 | 149 | self.dataOut.processingHeaderObj = self.dataIn.processingHeaderObj.copy() |
|
|
150 | 150 | self.dataOut.nProfiles = self.dataOut.nFFTPoints |
|
|
151 | 151 | #self.dataOut.nHeights = len(self.dataOut.heightList) |
|
|
152 | 152 | except Exception as e: |
|
|
153 | 153 | print("Error dataIn ",e) |
|
|
154 | 154 | |
|
|
155 | 155 | |
|
|
156 | 156 | |
|
|
157 | 157 | if shift_fft: |
|
|
158 | 158 | #desplaza a la derecha en el eje 2 determinadas posiciones |
|
|
159 | 159 | shift = int(self.dataOut.nFFTPoints/2) |
|
|
160 | 160 | self.dataOut.data_spc = numpy.roll(self.dataOut.data_spc, shift , axis=1) |
|
|
161 | 161 | |
|
|
162 | 162 | if self.dataOut.data_cspc is not None: |
|
|
163 | 163 | #desplaza a la derecha en el eje 2 determinadas posiciones |
|
|
164 | 164 | self.dataOut.data_cspc = numpy.roll(self.dataOut.data_cspc, shift, axis=1) |
|
|
165 | 165 | if pairsList: |
|
|
166 | 166 | self.__selectPairs(pairsList) |
|
|
167 | 167 | |
|
|
168 | 168 | |
|
|
169 | 169 | elif self.dataIn.type == "Voltage": |
|
|
170 | 170 | |
|
|
171 | 171 | self.dataOut.flagNoData = True |
|
|
172 | 172 | self.dataOut.radarControllerHeaderObj = self.dataIn.radarControllerHeaderObj.copy() |
|
|
173 | 173 | self.dataOut.processingHeaderObj = self.dataIn.processingHeaderObj.copy() |
|
|
174 | 174 | if nFFTPoints == None: |
|
|
175 | 175 | raise ValueError("This SpectraProc.run() need nFFTPoints input variable") |
|
|
176 | 176 | |
|
|
177 | 177 | if nProfiles == None: |
|
|
178 | 178 | nProfiles = nFFTPoints |
|
|
179 | 179 | |
|
|
180 | 180 | #if ippFactor == None: |
|
|
181 | 181 | # self.dataOut.ippFactor = 1 |
|
|
182 | 182 | if ippFactor == None: |
|
|
183 | 183 | self.dataOut.ippFactor = self.dataIn.ippFactor |
|
|
184 | 184 | else: |
|
|
185 | 185 | self.dataOut.ippFactor = ippFactor |
|
|
186 | 186 | |
|
|
187 | 187 | #print(" volts ch,prof, h: ", self.dataIn.data.shape) |
|
|
188 | 188 | if self.buffer is None: |
|
|
189 | 189 | if not zeroPad: |
|
|
190 | 190 | self.buffer = numpy.zeros((self.dataIn.nChannels, |
|
|
191 | 191 | nProfiles, |
|
|
192 | 192 | self.dataIn.nHeights), |
|
|
193 | 193 | dtype='complex') |
|
|
194 | 194 | zeroPoints = 0 |
|
|
195 | 195 | else: |
|
|
196 | 196 | self.buffer = numpy.zeros((self.dataIn.nChannels, |
|
|
197 | 197 | nFFTPoints+int(zeroPoints), |
|
|
198 | 198 | self.dataIn.nHeights), |
|
|
199 | 199 | dtype='complex') |
|
|
200 | 200 | |
|
|
201 | 201 | self.dataOut.nFFTPoints = nFFTPoints + int(zeroPoints) |
|
|
202 | 202 | |
|
|
203 | 203 | if self.dataIn.flagDataAsBlock: |
|
|
204 | 204 | nVoltProfiles = self.dataIn.data.shape[1] |
|
|
205 | 205 | zeroPoints = 0 |
|
|
206 | 206 | if nVoltProfiles == nProfiles or zeroPad: |
|
|
207 | 207 | self.buffer = self.dataIn.data.copy() |
|
|
208 | 208 | self.profIndex = nVoltProfiles |
|
|
209 | 209 | |
|
|
210 | 210 | elif nVoltProfiles < nProfiles: |
|
|
211 | 211 | |
|
|
212 | 212 | if self.profIndex == 0: |
|
|
213 | 213 | self.id_min = 0 |
|
|
214 | 214 | self.id_max = nVoltProfiles |
|
|
215 | 215 | |
|
|
216 | 216 | self.buffer[:, self.id_min:self.id_max, |
|
|
217 | 217 | :] = self.dataIn.data |
|
|
218 | 218 | self.profIndex += nVoltProfiles |
|
|
219 | 219 | self.id_min += nVoltProfiles |
|
|
220 | 220 | self.id_max += nVoltProfiles |
|
|
221 | 221 | else: |
|
|
222 | 222 | raise ValueError("The type object %s has %d profiles, it should just has %d profiles" % ( |
|
|
223 | 223 | self.dataIn.type, self.dataIn.data.shape[1], nProfiles)) |
|
|
224 | 224 | self.dataOut.flagNoData = True |
|
|
225 | 225 | else: |
|
|
226 | 226 | self.buffer[:, self.profIndex, :] = self.dataIn.data.copy() |
|
|
227 | 227 | self.profIndex += 1 |
|
|
228 | 228 | |
|
|
229 | 229 | if self.firstdatatime == None: |
|
|
230 | 230 | self.firstdatatime = self.dataIn.utctime |
|
|
231 | 231 | |
|
|
232 | 232 | if self.profIndex == nProfiles or (zeroPad and zeroPoints==0): |
|
|
233 | 233 | |
|
|
234 | 234 | self.__updateSpecFromVoltage() |
|
|
235 | 235 | |
|
|
236 | 236 | if pairsList == None: |
|
|
237 | 237 | self.dataOut.pairsList = [pair for pair in itertools.combinations(self.dataOut.channelList, 2)] |
|
|
238 | 238 | else: |
|
|
239 | 239 | self.dataOut.pairsList = pairsList |
|
|
240 | 240 | self.__getFft() |
|
|
241 | 241 | self.dataOut.flagNoData = False |
|
|
242 | 242 | self.firstdatatime = None |
|
|
243 | 243 | self.nsamplesFFT = self.profIndex |
|
|
244 | 244 | self.profIndex = 0 |
|
|
245 | 245 | |
|
|
246 | 246 | #update Processing Header: |
|
|
247 | 247 | self.dataOut.processingHeaderObj.dtype = "Spectra" |
|
|
248 | 248 | self.dataOut.processingHeaderObj.nFFTPoints = self.dataOut.nFFTPoints |
|
|
249 | 249 | self.dataOut.processingHeaderObj.nSamplesFFT = self.nsamplesFFT |
|
|
250 | 250 | self.dataOut.processingHeaderObj.nIncohInt = 1 |
|
|
251 | 251 | |
|
|
252 | 252 | |
|
|
253 | 253 | elif self.dataIn.type == "Parameters": |
|
|
254 | 254 | |
|
|
255 | 255 | self.dataOut.data_spc = self.dataIn.data_spc |
|
|
256 | 256 | self.dataOut.data_cspc = self.dataIn.data_cspc |
|
|
257 | 257 | self.dataOut.data_outlier = self.dataIn.data_outlier |
|
|
258 | 258 | self.dataOut.nProfiles = self.dataIn.nProfiles |
|
|
259 | 259 | self.dataOut.nIncohInt = self.dataIn.nIncohInt |
|
|
260 | 260 | self.dataOut.nFFTPoints = self.dataIn.nFFTPoints |
|
|
261 | 261 | self.dataOut.ippFactor = self.dataIn.ippFactor |
|
|
262 | 262 | self.dataOut.max_nIncohInt = self.dataIn.max_nIncohInt |
|
|
263 | 263 | self.dataOut.radarControllerHeaderObj = self.dataIn.radarControllerHeaderObj.copy() |
|
|
264 | 264 | self.dataOut.ProcessingHeader = self.dataIn.ProcessingHeader.copy() |
|
|
265 | 265 | self.dataOut.ippSeconds = self.dataIn.ippSeconds |
|
|
266 | 266 | self.dataOut.ipp = self.dataIn.ipp |
|
|
267 | 267 | #self.dataOut.abscissaList = self.dataIn.getVelRange(1) |
|
|
268 | 268 | #self.dataOut.spc_noise = self.dataIn.getNoise() |
|
|
269 | 269 | #self.dataOut.spc_range = (self.dataIn.getFreqRange(1) , self.dataIn.getAcfRange(1) , self.dataIn.getVelRange(1)) |
|
|
270 | 270 | # self.dataOut.normFactor = self.dataIn.normFactor |
|
|
271 | 271 | if hasattr(self.dataIn, 'channelList'): |
|
|
272 | 272 | self.dataOut.channelList = self.dataIn.channelList |
|
|
273 | 273 | if hasattr(self.dataIn, 'pairsList'): |
|
|
274 | 274 | self.dataOut.pairsList = self.dataIn.pairsList |
|
|
275 | 275 | self.dataOut.groupList = self.dataIn.pairsList |
|
|
276 | 276 | |
|
|
277 | 277 | self.dataOut.flagNoData = False |
|
|
278 | 278 | |
|
|
279 | 279 | if hasattr(self.dataIn, 'ChanDist'): #Distances of receiver channels |
|
|
280 | 280 | self.dataOut.ChanDist = self.dataIn.ChanDist |
|
|
281 | 281 | else: self.dataOut.ChanDist = None |
|
|
282 | 282 | |
|
|
283 | 283 | #if hasattr(self.dataIn, 'VelRange'): #Velocities range |
|
|
284 | 284 | # self.dataOut.VelRange = self.dataIn.VelRange |
|
|
285 | 285 | #else: self.dataOut.VelRange = None |
|
|
286 | 286 | |
|
|
287 | 287 | |
|
|
288 | 288 | |
|
|
289 | 289 | else: |
|
|
290 | 290 | raise ValueError("The type of input object {} is not valid".format( |
|
|
291 | 291 | self.dataIn.type)) |
|
|
292 | 292 | |
|
|
293 | 293 | |
|
|
294 | 294 | |
|
|
295 | 295 | |
|
|
296 | 296 | #print("spc proc Done", self.dataOut.data_spc.shape) |
|
|
297 | 297 | #print(self.dataOut.data_spc) |
|
|
298 | 298 | return |
|
|
299 | 299 | |
|
|
300 | 300 | def __selectPairs(self, pairsList): |
|
|
301 | 301 | |
|
|
302 | 302 | if not pairsList: |
|
|
303 | 303 | return |
|
|
304 | 304 | |
|
|
305 | 305 | pairs = [] |
|
|
306 | 306 | pairsIndex = [] |
|
|
307 | 307 | |
|
|
308 | 308 | for pair in pairsList: |
|
|
309 | 309 | if pair[0] not in self.dataOut.channelList or pair[1] not in self.dataOut.channelList: |
|
|
310 | 310 | continue |
|
|
311 | 311 | pairs.append(pair) |
|
|
312 | 312 | pairsIndex.append(pairs.index(pair)) |
|
|
313 | 313 | |
|
|
314 | 314 | self.dataOut.data_cspc = self.dataOut.data_cspc[pairsIndex] |
|
|
315 | 315 | self.dataOut.pairsList = pairs |
|
|
316 | 316 | |
|
|
317 | 317 | return |
|
|
318 | 318 | |
|
|
319 | 319 | def selectFFTs(self, minFFT, maxFFT ): |
|
|
320 | 320 | """ |
|
|
321 | 321 | Selecciona un bloque de datos en base a un grupo de valores de puntos FFTs segun el rango |
|
|
322 | 322 | minFFT<= FFT <= maxFFT |
|
|
323 | 323 | """ |
|
|
324 | 324 | |
|
|
325 | 325 | if (minFFT > maxFFT): |
|
|
326 | 326 | raise ValueError("Error selecting heights: Height range (%d,%d) is not valid" % (minFFT, maxFFT)) |
|
|
327 | 327 | |
|
|
328 | 328 | if (minFFT < self.dataOut.getFreqRange()[0]): |
|
|
329 | 329 | minFFT = self.dataOut.getFreqRange()[0] |
|
|
330 | 330 | |
|
|
331 | 331 | if (maxFFT > self.dataOut.getFreqRange()[-1]): |
|
|
332 | 332 | maxFFT = self.dataOut.getFreqRange()[-1] |
|
|
333 | 333 | |
|
|
334 | 334 | minIndex = 0 |
|
|
335 | 335 | maxIndex = 0 |
|
|
336 | 336 | FFTs = self.dataOut.getFreqRange() |
|
|
337 | 337 | |
|
|
338 | 338 | inda = numpy.where(FFTs >= minFFT) |
|
|
339 | 339 | indb = numpy.where(FFTs <= maxFFT) |
|
|
340 | 340 | |
|
|
341 | 341 | try: |
|
|
342 | 342 | minIndex = inda[0][0] |
|
|
343 | 343 | except: |
|
|
344 | 344 | minIndex = 0 |
|
|
345 | 345 | |
|
|
346 | 346 | try: |
|
|
347 | 347 | maxIndex = indb[0][-1] |
|
|
348 | 348 | except: |
|
|
349 | 349 | maxIndex = len(FFTs) |
|
|
350 | 350 | |
|
|
351 | 351 | self.selectFFTsByIndex(minIndex, maxIndex) |
|
|
352 | 352 | |
|
|
353 | 353 | return 1 |
|
|
354 | 354 | |
|
|
355 | 355 | def getBeaconSignal(self, tauindex=0, channelindex=0, hei_ref=None): |
|
|
356 | 356 | newheis = numpy.where( |
|
|
357 | 357 | self.dataOut.heightList > self.dataOut.radarControllerHeaderObj.Taus[tauindex]) |
|
|
358 | 358 | |
|
|
359 | 359 | if hei_ref != None: |
|
|
360 | 360 | newheis = numpy.where(self.dataOut.heightList > hei_ref) |
|
|
361 | 361 | |
|
|
362 | 362 | minIndex = min(newheis[0]) |
|
|
363 | 363 | maxIndex = max(newheis[0]) |
|
|
364 | 364 | data_spc = self.dataOut.data_spc[:, :, minIndex:maxIndex + 1] |
|
|
365 | 365 | heightList = self.dataOut.heightList[minIndex:maxIndex + 1] |
|
|
366 | 366 | |
|
|
367 | 367 | # determina indices |
|
|
368 | 368 | nheis = int(self.dataOut.radarControllerHeaderObj.txB / |
|
|
369 | 369 | (self.dataOut.heightList[1] - self.dataOut.heightList[0])) |
|
|
370 | 370 | avg_dB = 10 * \ |
|
|
371 | 371 | numpy.log10(numpy.sum(data_spc[channelindex, :, :], axis=0)) |
|
|
372 | 372 | beacon_dB = numpy.sort(avg_dB)[-nheis:] |
|
|
373 | 373 | beacon_heiIndexList = [] |
|
|
374 | 374 | for val in avg_dB.tolist(): |
|
|
375 | 375 | if val >= beacon_dB[0]: |
|
|
376 | 376 | beacon_heiIndexList.append(avg_dB.tolist().index(val)) |
|
|
377 | 377 | |
|
|
378 | 378 | #data_spc = data_spc[:,:,beacon_heiIndexList] |
|
|
379 | 379 | data_cspc = None |
|
|
380 | 380 | if self.dataOut.data_cspc is not None: |
|
|
381 | 381 | data_cspc = self.dataOut.data_cspc[:, :, minIndex:maxIndex + 1] |
|
|
382 | 382 | #data_cspc = data_cspc[:,:,beacon_heiIndexList] |
|
|
383 | 383 | |
|
|
384 | 384 | data_dc = None |
|
|
385 | 385 | if self.dataOut.data_dc is not None: |
|
|
386 | 386 | data_dc = self.dataOut.data_dc[:, minIndex:maxIndex + 1] |
|
|
387 | 387 | #data_dc = data_dc[:,beacon_heiIndexList] |
|
|
388 | 388 | |
|
|
389 | 389 | self.dataOut.data_spc = data_spc |
|
|
390 | 390 | self.dataOut.data_cspc = data_cspc |
|
|
391 | 391 | self.dataOut.data_dc = data_dc |
|
|
392 | 392 | self.dataOut.heightList = heightList |
|
|
393 | 393 | self.dataOut.beacon_heiIndexList = beacon_heiIndexList |
|
|
394 | 394 | |
|
|
395 | 395 | return 1 |
|
|
396 | 396 | |
|
|
397 | 397 | def selectFFTsByIndex(self, minIndex, maxIndex): |
|
|
398 | 398 | """ |
|
|
399 | 399 | |
|
|
400 | 400 | """ |
|
|
401 | 401 | |
|
|
402 | 402 | if (minIndex < 0) or (minIndex > maxIndex): |
|
|
403 | 403 | raise ValueError("Error selecting heights: Index range (%d,%d) is not valid" % (minIndex, maxIndex)) |
|
|
404 | 404 | |
|
|
405 | 405 | if (maxIndex >= self.dataOut.nProfiles): |
|
|
406 | 406 | maxIndex = self.dataOut.nProfiles-1 |
|
|
407 | 407 | |
|
|
408 | 408 | #Spectra |
|
|
409 | 409 | data_spc = self.dataOut.data_spc[:,minIndex:maxIndex+1,:] |
|
|
410 | 410 | |
|
|
411 | 411 | data_cspc = None |
|
|
412 | 412 | if self.dataOut.data_cspc is not None: |
|
|
413 | 413 | data_cspc = self.dataOut.data_cspc[:,minIndex:maxIndex+1,:] |
|
|
414 | 414 | |
|
|
415 | 415 | data_dc = None |
|
|
416 | 416 | if self.dataOut.data_dc is not None: |
|
|
417 | 417 | data_dc = self.dataOut.data_dc[minIndex:maxIndex+1,:] |
|
|
418 | 418 | |
|
|
419 | 419 | self.dataOut.data_spc = data_spc |
|
|
420 | 420 | self.dataOut.data_cspc = data_cspc |
|
|
421 | 421 | self.dataOut.data_dc = data_dc |
|
|
422 | 422 | |
|
|
423 | 423 | self.dataOut.ippSeconds = self.dataOut.ippSeconds*(self.dataOut.nFFTPoints / numpy.shape(data_cspc)[1]) |
|
|
424 | 424 | self.dataOut.nFFTPoints = numpy.shape(data_cspc)[1] |
|
|
425 | 425 | self.dataOut.profilesPerBlock = numpy.shape(data_cspc)[1] |
|
|
426 | 426 | |
|
|
427 | 427 | return 1 |
|
|
428 | 428 | |
|
|
429 | 429 | def getNoise(self, minHei=None, maxHei=None, minVel=None, maxVel=None): |
|
|
430 | 430 | # validacion de rango |
|
|
431 | 431 | if minHei == None: |
|
|
432 | 432 | minHei = self.dataOut.heightList[0] |
|
|
433 | 433 | |
|
|
434 | 434 | if maxHei == None: |
|
|
435 | 435 | maxHei = self.dataOut.heightList[-1] |
|
|
436 | 436 | |
|
|
437 | 437 | if (minHei < self.dataOut.heightList[0]) or (minHei > maxHei): |
|
|
438 | 438 | print('minHei: %.2f is out of the heights range' % (minHei)) |
|
|
439 | 439 | print('minHei is setting to %.2f' % (self.dataOut.heightList[0])) |
|
|
440 | 440 | minHei = self.dataOut.heightList[0] |
|
|
441 | 441 | |
|
|
442 | 442 | if (maxHei > self.dataOut.heightList[-1]) or (maxHei < minHei): |
|
|
443 | 443 | print('maxHei: %.2f is out of the heights range' % (maxHei)) |
|
|
444 | 444 | print('maxHei is setting to %.2f' % (self.dataOut.heightList[-1])) |
|
|
445 | 445 | maxHei = self.dataOut.heightList[-1] |
|
|
446 | 446 | |
|
|
447 | 447 | # validacion de velocidades |
|
|
448 | 448 | velrange = self.dataOut.getVelRange(1) |
|
|
449 | 449 | |
|
|
450 | 450 | if minVel == None: |
|
|
451 | 451 | minVel = velrange[0] |
|
|
452 | 452 | |
|
|
453 | 453 | if maxVel == None: |
|
|
454 | 454 | maxVel = velrange[-1] |
|
|
455 | 455 | |
|
|
456 | 456 | if (minVel < velrange[0]) or (minVel > maxVel): |
|
|
457 | 457 | print('minVel: %.2f is out of the velocity range' % (minVel)) |
|
|
458 | 458 | print('minVel is setting to %.2f' % (velrange[0])) |
|
|
459 | 459 | minVel = velrange[0] |
|
|
460 | 460 | |
|
|
461 | 461 | if (maxVel > velrange[-1]) or (maxVel < minVel): |
|
|
462 | 462 | print('maxVel: %.2f is out of the velocity range' % (maxVel)) |
|
|
463 | 463 | print('maxVel is setting to %.2f' % (velrange[-1])) |
|
|
464 | 464 | maxVel = velrange[-1] |
|
|
465 | 465 | |
|
|
466 | 466 | # seleccion de indices para rango |
|
|
467 | 467 | minIndex = 0 |
|
|
468 | 468 | maxIndex = 0 |
|
|
469 | 469 | heights = self.dataOut.heightList |
|
|
470 | 470 | |
|
|
471 | 471 | inda = numpy.where(heights >= minHei) |
|
|
472 | 472 | indb = numpy.where(heights <= maxHei) |
|
|
473 | 473 | |
|
|
474 | 474 | try: |
|
|
475 | 475 | minIndex = inda[0][0] |
|
|
476 | 476 | except: |
|
|
477 | 477 | minIndex = 0 |
|
|
478 | 478 | |
|
|
479 | 479 | try: |
|
|
480 | 480 | maxIndex = indb[0][-1] |
|
|
481 | 481 | except: |
|
|
482 | 482 | maxIndex = len(heights) |
|
|
483 | 483 | |
|
|
484 | 484 | if (minIndex < 0) or (minIndex > maxIndex): |
|
|
485 | 485 | raise ValueError("some value in (%d,%d) is not valid" % ( |
|
|
486 | 486 | minIndex, maxIndex)) |
|
|
487 | 487 | |
|
|
488 | 488 | if (maxIndex >= self.dataOut.nHeights): |
|
|
489 | 489 | maxIndex = self.dataOut.nHeights - 1 |
|
|
490 | 490 | |
|
|
491 | 491 | # seleccion de indices para velocidades |
|
|
492 | 492 | indminvel = numpy.where(velrange >= minVel) |
|
|
493 | 493 | indmaxvel = numpy.where(velrange <= maxVel) |
|
|
494 | 494 | try: |
|
|
495 | 495 | minIndexVel = indminvel[0][0] |
|
|
496 | 496 | except: |
|
|
497 | 497 | minIndexVel = 0 |
|
|
498 | 498 | |
|
|
499 | 499 | try: |
|
|
500 | 500 | maxIndexVel = indmaxvel[0][-1] |
|
|
501 | 501 | except: |
|
|
502 | 502 | maxIndexVel = len(velrange) |
|
|
503 | 503 | |
|
|
504 | 504 | # seleccion del espectro |
|
|
505 | 505 | data_spc = self.dataOut.data_spc[:, |
|
|
506 | 506 | minIndexVel:maxIndexVel + 1, minIndex:maxIndex + 1] |
|
|
507 | 507 | # estimacion de ruido |
|
|
508 | 508 | noise = numpy.zeros(self.dataOut.nChannels) |
|
|
509 | 509 | |
|
|
510 | 510 | for channel in range(self.dataOut.nChannels): |
|
|
511 | 511 | daux = data_spc[channel, :, :] |
|
|
512 | 512 | sortdata = numpy.sort(daux, axis=None) |
|
|
513 | 513 | noise[channel] = hildebrand_sekhon(sortdata, self.dataOut.nIncohInt) |
|
|
514 | 514 | |
|
|
515 | 515 | self.dataOut.noise_estimation = noise.copy() |
|
|
516 | 516 | |
|
|
517 | 517 | return 1 |
|
|
518 | 518 | |
|
|
519 | 519 | class removeDC(Operation): |
|
|
520 | 520 | |
|
|
521 | 521 | def run(self, dataOut, mode=2): |
|
|
522 | 522 | self.dataOut = dataOut |
|
|
523 | 523 | jspectra = self.dataOut.data_spc |
|
|
524 | 524 | jcspectra = self.dataOut.data_cspc |
|
|
525 | 525 | |
|
|
526 | 526 | num_chan = jspectra.shape[0] |
|
|
527 | 527 | num_hei = jspectra.shape[2] |
|
|
528 | 528 | |
|
|
529 | 529 | if jcspectra is not None: |
|
|
530 | 530 | jcspectraExist = True |
|
|
531 | 531 | num_pairs = jcspectra.shape[0] |
|
|
532 | 532 | else: |
|
|
533 | 533 | jcspectraExist = False |
|
|
534 | 534 | |
|
|
535 | 535 | freq_dc = int(jspectra.shape[1] / 2) |
|
|
536 | 536 | ind_vel = numpy.array([-2, -1, 1, 2]) + freq_dc |
|
|
537 | 537 | ind_vel = ind_vel.astype(int) |
|
|
538 | 538 | |
|
|
539 | 539 | if ind_vel[0] < 0: |
|
|
540 | 540 | ind_vel[list(range(0, 1))] = ind_vel[list(range(0, 1))] + self.num_prof |
|
|
541 | 541 | |
|
|
542 | 542 | if mode == 1: |
|
|
543 | 543 | jspectra[:, freq_dc, :] = ( |
|
|
544 | 544 | jspectra[:, ind_vel[1], :] + jspectra[:, ind_vel[2], :]) / 2 # CORRECCION |
|
|
545 | 545 | |
|
|
546 | 546 | if jcspectraExist: |
|
|
547 | 547 | jcspectra[:, freq_dc, :] = ( |
|
|
548 | 548 | jcspectra[:, ind_vel[1], :] + jcspectra[:, ind_vel[2], :]) / 2 |
|
|
549 | 549 | |
|
|
550 | 550 | if mode == 2: |
|
|
551 | 551 | |
|
|
552 | 552 | vel = numpy.array([-2, -1, 1, 2]) |
|
|
553 | 553 | xx = numpy.zeros([4, 4]) |
|
|
554 | 554 | |
|
|
555 | 555 | for fil in range(4): |
|
|
556 | 556 | xx[fil, :] = vel[fil]**numpy.asarray(list(range(4))) |
|
|
557 | 557 | |
|
|
558 | 558 | xx_inv = numpy.linalg.inv(xx) |
|
|
559 | 559 | xx_aux = xx_inv[0, :] |
|
|
560 | 560 | |
|
|
561 | 561 | for ich in range(num_chan): |
|
|
562 | 562 | yy = jspectra[ich, ind_vel, :] |
|
|
563 | 563 | jspectra[ich, freq_dc, :] = numpy.dot(xx_aux, yy) |
|
|
564 | 564 | |
|
|
565 | 565 | junkid = jspectra[ich, freq_dc, :] <= 0 |
|
|
566 | 566 | cjunkid = sum(junkid) |
|
|
567 | 567 | |
|
|
568 | 568 | if cjunkid.any(): |
|
|
569 | 569 | jspectra[ich, freq_dc, junkid.nonzero()] = ( |
|
|
570 | 570 | jspectra[ich, ind_vel[1], junkid] + jspectra[ich, ind_vel[2], junkid]) / 2 |
|
|
571 | 571 | |
|
|
572 | 572 | if jcspectraExist: |
|
|
573 | 573 | for ip in range(num_pairs): |
|
|
574 | 574 | yy = jcspectra[ip, ind_vel, :] |
|
|
575 | 575 | jcspectra[ip, freq_dc, :] = numpy.dot(xx_aux, yy) |
|
|
576 | 576 | |
|
|
577 | 577 | self.dataOut.data_spc = jspectra |
|
|
578 | 578 | self.dataOut.data_cspc = jcspectra |
|
|
579 | 579 | |
|
|
580 | 580 | return self.dataOut |
|
|
581 | 581 | |
|
|
582 | 582 | class getNoiseB(Operation): |
|
|
583 | 583 | |
|
|
584 | 584 | __slots__ =('offset','warnings', 'isConfig', 'minIndex','maxIndex','minIndexFFT','maxIndexFFT') |
|
|
585 | 585 | def __init__(self): |
|
|
586 | 586 | |
|
|
587 | 587 | Operation.__init__(self) |
|
|
588 | 588 | self.isConfig = False |
|
|
589 | 589 | |
|
|
590 | 590 | def setup(self, offset=None, minHei=None, maxHei=None,minVel=None, maxVel=None, minFreq= None, maxFreq=None, warnings=False): |
|
|
591 | 591 | |
|
|
592 | 592 | self.warnings = warnings |
|
|
593 | 593 | if minHei == None: |
|
|
594 | 594 | minHei = self.dataOut.heightList[0] |
|
|
595 | 595 | |
|
|
596 | 596 | if maxHei == None: |
|
|
597 | 597 | maxHei = self.dataOut.heightList[-1] |
|
|
598 | 598 | |
|
|
599 | 599 | if (minHei < self.dataOut.heightList[0]) or (minHei > maxHei): |
|
|
600 | 600 | if self.warnings: |
|
|
601 | 601 | print('minHei: %.2f is out of the heights range' % (minHei)) |
|
|
602 | 602 | print('minHei is setting to %.2f' % (self.dataOut.heightList[0])) |
|
|
603 | 603 | minHei = self.dataOut.heightList[0] |
|
|
604 | 604 | |
|
|
605 | 605 | if (maxHei > self.dataOut.heightList[-1]) or (maxHei < minHei): |
|
|
606 | 606 | if self.warnings: |
|
|
607 | 607 | print('maxHei: %.2f is out of the heights range' % (maxHei)) |
|
|
608 | 608 | print('maxHei is setting to %.2f' % (self.dataOut.heightList[-1])) |
|
|
609 | 609 | maxHei = self.dataOut.heightList[-1] |
|
|
610 | 610 | |
|
|
611 | 611 | |
|
|
612 | 612 | #indices relativos a los puntos de fft, puede ser de acuerdo a velocidad o frecuencia |
|
|
613 | 613 | minIndexFFT = 0 |
|
|
614 | 614 | maxIndexFFT = 0 |
|
|
615 | 615 | # validacion de velocidades |
|
|
616 | 616 | indminPoint = None |
|
|
617 | 617 | indmaxPoint = None |
|
|
618 | 618 | if self.dataOut.type == 'Spectra': |
|
|
619 | 619 | if minVel == None and maxVel == None : |
|
|
620 | 620 | |
|
|
621 | 621 | freqrange = self.dataOut.getFreqRange(1) |
|
|
622 | 622 | |
|
|
623 | 623 | if minFreq == None: |
|
|
624 | 624 | minFreq = freqrange[0] |
|
|
625 | 625 | |
|
|
626 | 626 | if maxFreq == None: |
|
|
627 | 627 | maxFreq = freqrange[-1] |
|
|
628 | 628 | |
|
|
629 | 629 | if (minFreq < freqrange[0]) or (minFreq > maxFreq): |
|
|
630 | 630 | if self.warnings: |
|
|
631 | 631 | print('minFreq: %.2f is out of the frequency range' % (minFreq)) |
|
|
632 | 632 | print('minFreq is setting to %.2f' % (freqrange[0])) |
|
|
633 | 633 | minFreq = freqrange[0] |
|
|
634 | 634 | |
|
|
635 | 635 | if (maxFreq > freqrange[-1]) or (maxFreq < minFreq): |
|
|
636 | 636 | if self.warnings: |
|
|
637 | 637 | print('maxFreq: %.2f is out of the frequency range' % (maxFreq)) |
|
|
638 | 638 | print('maxFreq is setting to %.2f' % (freqrange[-1])) |
|
|
639 | 639 | maxFreq = freqrange[-1] |
|
|
640 | 640 | |
|
|
641 | 641 | indminPoint = numpy.where(freqrange >= minFreq) |
|
|
642 | 642 | indmaxPoint = numpy.where(freqrange <= maxFreq) |
|
|
643 | 643 | |
|
|
644 | 644 | else: |
|
|
645 | 645 | |
|
|
646 | 646 | velrange = self.dataOut.getVelRange(1) |
|
|
647 | 647 | |
|
|
648 | 648 | if minVel == None: |
|
|
649 | 649 | minVel = velrange[0] |
|
|
650 | 650 | |
|
|
651 | 651 | if maxVel == None: |
|
|
652 | 652 | maxVel = velrange[-1] |
|
|
653 | 653 | |
|
|
654 | 654 | if (minVel < velrange[0]) or (minVel > maxVel): |
|
|
655 | 655 | if self.warnings: |
|
|
656 | 656 | print('minVel: %.2f is out of the velocity range' % (minVel)) |
|
|
657 | 657 | print('minVel is setting to %.2f' % (velrange[0])) |
|
|
658 | 658 | minVel = velrange[0] |
|
|
659 | 659 | |
|
|
660 | 660 | if (maxVel > velrange[-1]) or (maxVel < minVel): |
|
|
661 | 661 | if self.warnings: |
|
|
662 | 662 | print('maxVel: %.2f is out of the velocity range' % (maxVel)) |
|
|
663 | 663 | print('maxVel is setting to %.2f' % (velrange[-1])) |
|
|
664 | 664 | maxVel = velrange[-1] |
|
|
665 | 665 | |
|
|
666 | 666 | indminPoint = numpy.where(velrange >= minVel) |
|
|
667 | 667 | indmaxPoint = numpy.where(velrange <= maxVel) |
|
|
668 | 668 | |
|
|
669 | 669 | |
|
|
670 | 670 | # seleccion de indices para rango |
|
|
671 | 671 | minIndex = 0 |
|
|
672 | 672 | maxIndex = 0 |
|
|
673 | 673 | heights = self.dataOut.heightList |
|
|
674 | 674 | |
|
|
675 | 675 | inda = numpy.where(heights >= minHei) |
|
|
676 | 676 | indb = numpy.where(heights <= maxHei) |
|
|
677 | 677 | |
|
|
678 | 678 | try: |
|
|
679 | 679 | minIndex = inda[0][0] |
|
|
680 | 680 | except: |
|
|
681 | 681 | minIndex = 0 |
|
|
682 | 682 | |
|
|
683 | 683 | try: |
|
|
684 | 684 | maxIndex = indb[0][-1] |
|
|
685 | 685 | except: |
|
|
686 | 686 | maxIndex = len(heights) |
|
|
687 | 687 | |
|
|
688 | 688 | if (minIndex < 0) or (minIndex > maxIndex): |
|
|
689 | 689 | raise ValueError("some value in (%d,%d) is not valid" % ( |
|
|
690 | 690 | minIndex, maxIndex)) |
|
|
691 | 691 | |
|
|
692 | 692 | if (maxIndex >= self.dataOut.nHeights): |
|
|
693 | 693 | maxIndex = self.dataOut.nHeights - 1 |
|
|
694 | 694 | #############################################################3 |
|
|
695 | 695 | # seleccion de indices para velocidades |
|
|
696 | 696 | if self.dataOut.type == 'Spectra': |
|
|
697 | 697 | try: |
|
|
698 | 698 | minIndexFFT = indminPoint[0][0] |
|
|
699 | 699 | except: |
|
|
700 | 700 | minIndexFFT = 0 |
|
|
701 | 701 | |
|
|
702 | 702 | try: |
|
|
703 | 703 | maxIndexFFT = indmaxPoint[0][-1] |
|
|
704 | 704 | except: |
|
|
705 | 705 | maxIndexFFT = len( self.dataOut.getFreqRange(1)) |
|
|
706 | 706 | |
|
|
707 | 707 | self.minIndex, self.maxIndex, self.minIndexFFT, self.maxIndexFFT = minIndex, maxIndex, minIndexFFT, maxIndexFFT |
|
|
708 | 708 | self.isConfig = True |
|
|
709 | 709 | self.offset = 1 |
|
|
710 | 710 | if offset!=None: |
|
|
711 | 711 | self.offset = 10**(offset/10) |
|
|
712 | 712 | #print("config getNoiseB Done") |
|
|
713 | 713 | |
|
|
714 | 714 | def run(self, dataOut, offset=None, minHei=None, maxHei=None,minVel=None, maxVel=None, minFreq= None, maxFreq=None, warnings=False): |
|
|
715 | 715 | self.dataOut = dataOut |
|
|
716 | 716 | |
|
|
717 | 717 | if not self.isConfig: |
|
|
718 | 718 | self.setup(offset, minHei, maxHei,minVel, maxVel, minFreq, maxFreq, warnings) |
|
|
719 | 719 | |
|
|
720 | 720 | self.dataOut.noise_estimation = None |
|
|
721 | 721 | noise = None |
|
|
722 | 722 | #print("data type: ",self.dataOut.type, self.dataOut.nIncohInt, self.dataOut.max_nIncohInt) |
|
|
723 | 723 | if self.dataOut.type == 'Voltage': |
|
|
724 | 724 | noise = self.dataOut.getNoise(ymin_index=self.minIndex, ymax_index=self.maxIndex) |
|
|
725 | 725 | #print(minIndex, maxIndex,minIndexVel, maxIndexVel) |
|
|
726 | 726 | elif self.dataOut.type == 'Spectra': |
|
|
727 | 727 | #print(self.dataOut.nChannels, self.minIndex, self.maxIndex,self.minIndexFFT, self.maxIndexFFT, self.dataOut.max_nIncohInt, self.dataOut.nIncohInt) |
|
|
728 | 728 | noise = numpy.zeros( self.dataOut.nChannels) |
|
|
729 | 729 | norm = 1 |
|
|
730 | 730 | |
|
|
731 | 731 | for channel in range( self.dataOut.nChannels): |
|
|
732 | 732 | if not hasattr(self.dataOut.nIncohInt,'__len__'): |
|
|
733 | 733 | norm = 1 |
|
|
734 | 734 | else: |
|
|
735 | 735 | norm = self.dataOut.max_nIncohInt[channel]/self.dataOut.nIncohInt[channel, self.minIndex:self.maxIndex] |
|
|
736 | 736 | |
|
|
737 | 737 | #print("norm nIncoh: ", norm ,self.dataOut.data_spc.shape, self.dataOut.max_nIncohInt) |
|
|
738 | 738 | daux = self.dataOut.data_spc[channel,self.minIndexFFT:self.maxIndexFFT, self.minIndex:self.maxIndex] |
|
|
739 | 739 | daux = numpy.multiply(daux, norm) |
|
|
740 | 740 | #print("offset: ", self.offset, 10*numpy.log10(self.offset)) |
|
|
741 | 741 | # noise[channel] = self.getNoiseByMean(daux)/self.offset |
|
|
742 | 742 | #print(daux.shape, daux) |
|
|
743 | 743 | #noise[channel] = self.getNoiseByHS(daux, self.dataOut.max_nIncohInt)/self.offset |
|
|
744 | 744 | sortdata = numpy.sort(daux, axis=None) |
|
|
745 | 745 | |
|
|
746 | 746 | noise[channel] = _noise.hildebrand_sekhon(sortdata, self.dataOut.max_nIncohInt[channel])/self.offset |
|
|
747 | 747 | #print("noise shape", noise[channel], self.name) |
|
|
748 | 748 | |
|
|
749 | 749 | #noise = self.dataOut.getNoise(xmin_index=self.minIndexFFT, xmax_index=self.maxIndexFFT, ymin_index=self.minIndex, ymax_index=self.maxIndex) |
|
|
750 | 750 | else: |
|
|
751 | 751 | noise = self.dataOut.getNoise(xmin_index=self.minIndexFFT, xmax_index=self.maxIndexFFT, ymin_index=self.minIndex, ymax_index=self.maxIndex) |
|
|
752 | 752 | |
|
|
753 | 753 | self.dataOut.noise_estimation = noise.copy() # dataOut.noise |
|
|
754 | 754 | #print("2: ",10*numpy.log10(self.dataOut.noise_estimation/64)) |
|
|
755 | 755 | #print("2: ",self.dataOut.noise_estimation) |
|
|
756 | 756 | #print(self.dataOut.flagNoData) |
|
|
757 | 757 | #print("getNoise Done", 10*numpy.log10(noise)) |
|
|
758 | 758 | return self.dataOut |
|
|
759 | 759 | |
|
|
760 | 760 | def getNoiseByMean(self,data): |
|
|
761 | 761 | #data debe estar ordenado |
|
|
762 | 762 | data = numpy.mean(data,axis=1) |
|
|
763 | 763 | sortdata = numpy.sort(data, axis=None) |
|
|
764 | 764 | #sortID=data.argsort() |
|
|
765 | 765 | #print(data.shape) |
|
|
766 | 766 | |
|
|
767 | 767 | pnoise = None |
|
|
768 | 768 | j = 0 |
|
|
769 | 769 | |
|
|
770 | 770 | mean = numpy.mean(sortdata) |
|
|
771 | 771 | min = numpy.min(sortdata) |
|
|
772 | 772 | delta = mean - min |
|
|
773 | 773 | indexes = numpy.where(sortdata > (mean+delta))[0] #only array of indexes |
|
|
774 | 774 | #print(len(indexes)) |
|
|
775 | 775 | if len(indexes)==0: |
|
|
776 | 776 | pnoise = numpy.mean(sortdata) |
|
|
777 | 777 | else: |
|
|
778 | 778 | j = indexes[0] |
|
|
779 | 779 | pnoise = numpy.mean(sortdata[0:j]) |
|
|
780 | 780 | |
|
|
781 | 781 | # from matplotlib import pyplot as plt |
|
|
782 | 782 | # plt.plot(sortdata) |
|
|
783 | 783 | # plt.vlines(j,(pnoise-delta),(pnoise+delta), color='r') |
|
|
784 | 784 | # plt.show() |
|
|
785 | 785 | #print("noise: ", 10*numpy.log10(pnoise)) |
|
|
786 | 786 | return pnoise |
|
|
787 | 787 | |
|
|
788 | 788 | def getNoiseByHS(self,data, navg): |
|
|
789 | 789 | #data debe estar ordenado |
|
|
790 | 790 | #data = numpy.mean(data,axis=1) |
|
|
791 | 791 | sortdata = numpy.sort(data, axis=None) |
|
|
792 | 792 | |
|
|
793 | 793 | lenOfData = len(sortdata) |
|
|
794 | 794 | nums_min = lenOfData*0.2 |
|
|
795 | 795 | |
|
|
796 | 796 | if nums_min <= 5: |
|
|
797 | 797 | |
|
|
798 | 798 | nums_min = 5 |
|
|
799 | 799 | |
|
|
800 | 800 | sump = 0. |
|
|
801 | 801 | sumq = 0. |
|
|
802 | 802 | |
|
|
803 | 803 | j = 0 |
|
|
804 | 804 | cont = 1 |
|
|
805 | 805 | |
|
|
806 | 806 | while((cont == 1)and(j < lenOfData)): |
|
|
807 | 807 | |
|
|
808 | 808 | sump += sortdata[j] |
|
|
809 | 809 | sumq += sortdata[j]**2 |
|
|
810 | 810 | #sumq -= sump**2 |
|
|
811 | 811 | if j > nums_min: |
|
|
812 | 812 | rtest = float(j)/(j-1) + 1.0/navg |
|
|
813 | 813 | #if ((sumq*j) > (sump**2)): |
|
|
814 | 814 | if ((sumq*j) > (rtest*sump**2)): |
|
|
815 | 815 | j = j - 1 |
|
|
816 | 816 | sump = sump - sortdata[j] |
|
|
817 | 817 | sumq = sumq - sortdata[j]**2 |
|
|
818 | 818 | cont = 0 |
|
|
819 | 819 | |
|
|
820 | 820 | j += 1 |
|
|
821 | 821 | |
|
|
822 | 822 | lnoise = sump / j |
|
|
823 | 823 | |
|
|
824 | 824 | return lnoise |
|
|
825 | 825 | |
|
|
826 | 826 | |
|
|
827 | 827 | |
|
|
828 | 828 | def fit_func( x, a0, a1, a2): #, a3, a4, a5): |
|
|
829 | 829 | z = (x - a1) / a2 |
|
|
830 | 830 | y = a0 * numpy.exp(-z**2 / a2) #+ a3 + a4 * x + a5 * x**2 |
|
|
831 | 831 | return y |
|
|
832 | 832 | |
|
|
833 | 833 | |
|
|
834 | 834 | # class CleanRayleigh(Operation): |
|
|
835 | 835 | # |
|
|
836 | 836 | # def __init__(self): |
|
|
837 | 837 | # |
|
|
838 | 838 | # Operation.__init__(self) |
|
|
839 | 839 | # self.i=0 |
|
|
840 | 840 | # self.isConfig = False |
|
|
841 | 841 | # self.__dataReady = False |
|
|
842 | 842 | # self.__profIndex = 0 |
|
|
843 | 843 | # self.byTime = False |
|
|
844 | 844 | # self.byProfiles = False |
|
|
845 | 845 | # |
|
|
846 | 846 | # self.bloques = None |
|
|
847 | 847 | # self.bloque0 = None |
|
|
848 | 848 | # |
|
|
849 | 849 | # self.index = 0 |
|
|
850 | 850 | # |
|
|
851 | 851 | # self.buffer = 0 |
|
|
852 | 852 | # self.buffer2 = 0 |
|
|
853 | 853 | # self.buffer3 = 0 |
|
|
854 | 854 | # |
|
|
855 | 855 | # |
|
|
856 | 856 | # def setup(self,dataOut,min_hei,max_hei,n, timeInterval,factor_stdv): |
|
|
857 | 857 | # |
|
|
858 | 858 | # self.nChannels = dataOut.nChannels |
|
|
859 | 859 | # self.nProf = dataOut.nProfiles |
|
|
860 | 860 | # self.nPairs = dataOut.data_cspc.shape[0] |
|
|
861 | 861 | # self.pairsArray = numpy.array(dataOut.pairsList) |
|
|
862 | 862 | # self.spectra = dataOut.data_spc |
|
|
863 | 863 | # self.cspectra = dataOut.data_cspc |
|
|
864 | 864 | # self.heights = dataOut.heightList #alturas totales |
|
|
865 | 865 | # self.nHeights = len(self.heights) |
|
|
866 | 866 | # self.min_hei = min_hei |
|
|
867 | 867 | # self.max_hei = max_hei |
|
|
868 | 868 | # if (self.min_hei == None): |
|
|
869 | 869 | # self.min_hei = 0 |
|
|
870 | 870 | # if (self.max_hei == None): |
|
|
871 | 871 | # self.max_hei = dataOut.heightList[-1] |
|
|
872 | 872 | # self.hval = ((self.max_hei>=self.heights) & (self.heights >= self.min_hei)).nonzero() |
|
|
873 | 873 | # self.heightsClean = self.heights[self.hval] #alturas filtradas |
|
|
874 | 874 | # self.hval = self.hval[0] # forma (N,), an solo N elementos -> Indices de alturas |
|
|
875 | 875 | # self.nHeightsClean = len(self.heightsClean) |
|
|
876 | 876 | # self.channels = dataOut.channelList |
|
|
877 | 877 | # self.nChan = len(self.channels) |
|
|
878 | 878 | # self.nIncohInt = dataOut.nIncohInt |
|
|
879 | 879 | # self.__initime = dataOut.utctime |
|
|
880 | 880 | # self.maxAltInd = self.hval[-1]+1 |
|
|
881 | 881 | # self.minAltInd = self.hval[0] |
|
|
882 | 882 | # |
|
|
883 | 883 | # self.crosspairs = dataOut.pairsList |
|
|
884 | 884 | # self.nPairs = len(self.crosspairs) |
|
|
885 | 885 | # self.normFactor = dataOut.normFactor |
|
|
886 | 886 | # self.nFFTPoints = dataOut.nFFTPoints |
|
|
887 | 887 | # self.ippSeconds = dataOut.ippSeconds |
|
|
888 | 888 | # self.currentTime = self.__initime |
|
|
889 | 889 | # self.pairsArray = numpy.array(dataOut.pairsList) |
|
|
890 | 890 | # self.factor_stdv = factor_stdv |
|
|
891 | 891 | # |
|
|
892 | 892 | # if n != None : |
|
|
893 | 893 | # self.byProfiles = True |
|
|
894 | 894 | # self.nIntProfiles = n |
|
|
895 | 895 | # else: |
|
|
896 | 896 | # self.__integrationtime = timeInterval |
|
|
897 | 897 | # |
|
|
898 | 898 | # self.__dataReady = False |
|
|
899 | 899 | # self.isConfig = True |
|
|
900 | 900 | # |
|
|
901 | 901 | # |
|
|
902 | 902 | # |
|
|
903 | 903 | # def run(self, dataOut,min_hei=None,max_hei=None, n=None, timeInterval=10,factor_stdv=2.5): |
|
|
904 | 904 | # #print("runing cleanRayleigh") |
|
|
905 | 905 | # if not self.isConfig : |
|
|
906 | 906 | # |
|
|
907 | 907 | # self.setup(dataOut, min_hei,max_hei,n,timeInterval,factor_stdv) |
|
|
908 | 908 | # |
|
|
909 | 909 | # tini=dataOut.utctime |
|
|
910 | 910 | # |
|
|
911 | 911 | # if self.byProfiles: |
|
|
912 | 912 | # if self.__profIndex == self.nIntProfiles: |
|
|
913 | 913 | # self.__dataReady = True |
|
|
914 | 914 | # else: |
|
|
915 | 915 | # if (tini - self.__initime) >= self.__integrationtime: |
|
|
916 | 916 | # |
|
|
917 | 917 | # self.__dataReady = True |
|
|
918 | 918 | # self.__initime = tini |
|
|
919 | 919 | # |
|
|
920 | 920 | # #if (tini.tm_min % 2) == 0 and (tini.tm_sec < 5 and self.fint==0): |
|
|
921 | 921 | # |
|
|
922 | 922 | # if self.__dataReady: |
|
|
923 | 923 | # |
|
|
924 | 924 | # self.__profIndex = 0 |
|
|
925 | 925 | # jspc = self.buffer |
|
|
926 | 926 | # jcspc = self.buffer2 |
|
|
927 | 927 | # #jnoise = self.buffer3 |
|
|
928 | 928 | # self.buffer = dataOut.data_spc |
|
|
929 | 929 | # self.buffer2 = dataOut.data_cspc |
|
|
930 | 930 | # #self.buffer3 = dataOut.noise |
|
|
931 | 931 | # self.currentTime = dataOut.utctime |
|
|
932 | 932 | # if numpy.any(jspc) : |
|
|
933 | 933 | # #print( jspc.shape, jcspc.shape) |
|
|
934 | 934 | # jspc = numpy.reshape(jspc,(int(len(jspc)/self.nChannels),self.nChannels,self.nFFTPoints,self.nHeights)) |
|
|
935 | 935 | # try: |
|
|
936 | 936 | # jcspc= numpy.reshape(jcspc,(int(len(jcspc)/self.nPairs),self.nPairs,self.nFFTPoints,self.nHeights)) |
|
|
937 | 937 | # except: |
|
|
938 | 938 | # print("no cspc") |
|
|
939 | 939 | # self.__dataReady = False |
|
|
940 | 940 | # #print( jspc.shape, jcspc.shape) |
|
|
941 | 941 | # dataOut.flagNoData = False |
|
|
942 | 942 | # else: |
|
|
943 | 943 | # dataOut.flagNoData = True |
|
|
944 | 944 | # self.__dataReady = False |
|
|
945 | 945 | # return dataOut |
|
|
946 | 946 | # else: |
|
|
947 | 947 | # #print( len(self.buffer)) |
|
|
948 | 948 | # if numpy.any(self.buffer): |
|
|
949 | 949 | # self.buffer = numpy.concatenate((self.buffer,dataOut.data_spc), axis=0) |
|
|
950 | 950 | # try: |
|
|
951 | 951 | # self.buffer2 = numpy.concatenate((self.buffer2,dataOut.data_cspc), axis=0) |
|
|
952 | 952 | # self.buffer3 += dataOut.data_dc |
|
|
953 | 953 | # except: |
|
|
954 | 954 | # pass |
|
|
955 | 955 | # else: |
|
|
956 | 956 | # self.buffer = dataOut.data_spc |
|
|
957 | 957 | # self.buffer2 = dataOut.data_cspc |
|
|
958 | 958 | # self.buffer3 = dataOut.data_dc |
|
|
959 | 959 | # #print self.index, self.fint |
|
|
960 | 960 | # #print self.buffer2.shape |
|
|
961 | 961 | # dataOut.flagNoData = True ## NOTE: ?? revisar LUEGO |
|
|
962 | 962 | # self.__profIndex += 1 |
|
|
963 | 963 | # return dataOut ## NOTE: REV |
|
|
964 | 964 | # |
|
|
965 | 965 | # |
|
|
966 | 966 | # #index = tini.tm_hour*12+tini.tm_min/5 |
|
|
967 | 967 | # ''' |
|
|
968 | 968 | # #REVISAR |
|
|
969 | 969 | # ''' |
|
|
970 | 970 | # # jspc = jspc/self.nFFTPoints/self.normFactor |
|
|
971 | 971 | # # jcspc = jcspc/self.nFFTPoints/self.normFactor |
|
|
972 | 972 | # |
|
|
973 | 973 | # |
|
|
974 | 974 | # |
|
|
975 | 975 | # tmp_spectra,tmp_cspectra = self.cleanRayleigh(dataOut,jspc,jcspc,self.factor_stdv) |
|
|
976 | 976 | # dataOut.data_spc = tmp_spectra |
|
|
977 | 977 | # dataOut.data_cspc = tmp_cspectra |
|
|
978 | 978 | # |
|
|
979 | 979 | # #dataOut.data_spc,dataOut.data_cspc = self.cleanRayleigh(dataOut,jspc,jcspc,self.factor_stdv) |
|
|
980 | 980 | # |
|
|
981 | 981 | # dataOut.data_dc = self.buffer3 |
|
|
982 | 982 | # dataOut.nIncohInt *= self.nIntProfiles |
|
|
983 | 983 | # dataOut.max_nIncohInt = self.nIntProfiles |
|
|
984 | 984 | # dataOut.utctime = self.currentTime #tiempo promediado |
|
|
985 | 985 | # #print("Time: ",time.localtime(dataOut.utctime)) |
|
|
986 | 986 | # # dataOut.data_spc = sat_spectra |
|
|
987 | 987 | # # dataOut.data_cspc = sat_cspectra |
|
|
988 | 988 | # self.buffer = 0 |
|
|
989 | 989 | # self.buffer2 = 0 |
|
|
990 | 990 | # self.buffer3 = 0 |
|
|
991 | 991 | # |
|
|
992 | 992 | # return dataOut |
|
|
993 | 993 | # |
|
|
994 | 994 | # def cleanRayleigh(self,dataOut,spectra,cspectra,factor_stdv): |
|
|
995 | 995 | # print("OP cleanRayleigh") |
|
|
996 | 996 | # #import matplotlib.pyplot as plt |
|
|
997 | 997 | # #for k in range(149): |
|
|
998 | 998 | # #channelsProcssd = [] |
|
|
999 | 999 | # #channelA_ok = False |
|
|
1000 | 1000 | # #rfunc = cspectra.copy() #self.bloques |
|
|
1001 | 1001 | # rfunc = spectra.copy() |
|
|
1002 | 1002 | # #rfunc = cspectra |
|
|
1003 | 1003 | # #val_spc = spectra*0.0 #self.bloque0*0.0 |
|
|
1004 | 1004 | # #val_cspc = cspectra*0.0 #self.bloques*0.0 |
|
|
1005 | 1005 | # #in_sat_spectra = spectra.copy() #self.bloque0 |
|
|
1006 | 1006 | # #in_sat_cspectra = cspectra.copy() #self.bloques |
|
|
1007 | 1007 | # |
|
|
1008 | 1008 | # |
|
|
1009 | 1009 | # ###ONLY FOR TEST: |
|
|
1010 | 1010 | # raxs = math.ceil(math.sqrt(self.nPairs)) |
|
|
1011 | 1011 | # if raxs == 0: |
|
|
1012 | 1012 | # raxs = 1 |
|
|
1013 | 1013 | # caxs = math.ceil(self.nPairs/raxs) |
|
|
1014 | 1014 | # if self.nPairs <4: |
|
|
1015 | 1015 | # raxs = 2 |
|
|
1016 | 1016 | # caxs = 2 |
|
|
1017 | 1017 | # #print(raxs, caxs) |
|
|
1018 | 1018 | # fft_rev = 14 #nFFT to plot |
|
|
1019 | 1019 | # hei_rev = ((self.heights >= 550) & (self.heights <= 551)).nonzero() #hei to plot |
|
|
1020 | 1020 | # hei_rev = hei_rev[0] |
|
|
1021 | 1021 | # #print(hei_rev) |
|
|
1022 | 1022 | # |
|
|
1023 | 1023 | # #print numpy.absolute(rfunc[:,0,0,14]) |
|
|
1024 | 1024 | # |
|
|
1025 | 1025 | # gauss_fit, covariance = None, None |
|
|
1026 | 1026 | # for ih in range(self.minAltInd,self.maxAltInd): |
|
|
1027 | 1027 | # for ifreq in range(self.nFFTPoints): |
|
|
1028 | 1028 | # ''' |
|
|
1029 | 1029 | # ###ONLY FOR TEST: |
|
|
1030 | 1030 | # if ifreq ==fft_rev and ih==hei_rev: #TO VIEW A SIGNLE FREQUENCY |
|
|
1031 | 1031 | # fig, axs = plt.subplots(raxs, caxs) |
|
|
1032 | 1032 | # fig2, axs2 = plt.subplots(raxs, caxs) |
|
|
1033 | 1033 | # col_ax = 0 |
|
|
1034 | 1034 | # row_ax = 0 |
|
|
1035 | 1035 | # ''' |
|
|
1036 | 1036 | # #print(self.nPairs) |
|
|
1037 | 1037 | # for ii in range(self.nChan): #PARES DE CANALES SELF y CROSS |
|
|
1038 | 1038 | # # if self.crosspairs[ii][1]-self.crosspairs[ii][0] > 1: # APLICAR SOLO EN PARES CONTIGUOS |
|
|
1039 | 1039 | # # continue |
|
|
1040 | 1040 | # # if not self.crosspairs[ii][0] in channelsProcssd: |
|
|
1041 | 1041 | # # channelA_ok = True |
|
|
1042 | 1042 | # #print("pair: ",self.crosspairs[ii]) |
|
|
1043 | 1043 | # ''' |
|
|
1044 | 1044 | # ###ONLY FOR TEST: |
|
|
1045 | 1045 | # if (col_ax%caxs==0 and col_ax!=0 and self.nPairs !=1): |
|
|
1046 | 1046 | # col_ax = 0 |
|
|
1047 | 1047 | # row_ax += 1 |
|
|
1048 | 1048 | # ''' |
|
|
1049 | 1049 | # func2clean = 10*numpy.log10(numpy.absolute(rfunc[:,ii,ifreq,ih])) #Potencia? |
|
|
1050 | 1050 | # #print(func2clean.shape) |
|
|
1051 | 1051 | # val = (numpy.isfinite(func2clean)==True).nonzero() |
|
|
1052 | 1052 | # |
|
|
1053 | 1053 | # if len(val)>0: #limitador |
|
|
1054 | 1054 | # min_val = numpy.around(numpy.amin(func2clean)-2) #> (-40) |
|
|
1055 | 1055 | # if min_val <= -40 : |
|
|
1056 | 1056 | # min_val = -40 |
|
|
1057 | 1057 | # max_val = numpy.around(numpy.amax(func2clean)+2) #< 200 |
|
|
1058 | 1058 | # if max_val >= 200 : |
|
|
1059 | 1059 | # max_val = 200 |
|
|
1060 | 1060 | # #print min_val, max_val |
|
|
1061 | 1061 | # step = 1 |
|
|
1062 | 1062 | # #print("Getting bins and the histogram") |
|
|
1063 | 1063 | # x_dist = min_val + numpy.arange(1 + ((max_val-(min_val))/step))*step |
|
|
1064 | 1064 | # y_dist,binstep = numpy.histogram(func2clean,bins=range(int(min_val),int(max_val+2),step)) |
|
|
1065 | 1065 | # #print(len(y_dist),len(binstep[:-1])) |
|
|
1066 | 1066 | # #print(row_ax,col_ax, " ..") |
|
|
1067 | 1067 | # #print(self.pairsArray[ii][0],self.pairsArray[ii][1]) |
|
|
1068 | 1068 | # mean = numpy.sum(x_dist * y_dist) / numpy.sum(y_dist) |
|
|
1069 | 1069 | # sigma = numpy.sqrt(numpy.sum(y_dist * (x_dist - mean)**2) / numpy.sum(y_dist)) |
|
|
1070 | 1070 | # parg = [numpy.amax(y_dist),mean,sigma] |
|
|
1071 | 1071 | # |
|
|
1072 | 1072 | # newY = None |
|
|
1073 | 1073 | # |
|
|
1074 | 1074 | # try : |
|
|
1075 | 1075 | # gauss_fit, covariance = curve_fit(fit_func, x_dist, y_dist,p0=parg) |
|
|
1076 | 1076 | # mode = gauss_fit[1] |
|
|
1077 | 1077 | # stdv = gauss_fit[2] |
|
|
1078 | 1078 | # #print(" FIT OK",gauss_fit) |
|
|
1079 | 1079 | # ''' |
|
|
1080 | 1080 | # ###ONLY FOR TEST: |
|
|
1081 | 1081 | # if ifreq ==fft_rev and ih==hei_rev: #TO VIEW A SIGNLE FREQUENCY |
|
|
1082 | 1082 | # newY = fit_func(x_dist,gauss_fit[0],gauss_fit[1],gauss_fit[2]) |
|
|
1083 | 1083 | # axs[row_ax,col_ax].plot(binstep[:-1],y_dist,color='green') |
|
|
1084 | 1084 | # axs[row_ax,col_ax].plot(binstep[:-1],newY,color='red') |
|
|
1085 | 1085 | # axs[row_ax,col_ax].set_title("CH "+str(self.channels[ii])) |
|
|
1086 | 1086 | # ''' |
|
|
1087 | 1087 | # except: |
|
|
1088 | 1088 | # mode = mean |
|
|
1089 | 1089 | # stdv = sigma |
|
|
1090 | 1090 | # #print("FIT FAIL") |
|
|
1091 | 1091 | # #continue |
|
|
1092 | 1092 | # |
|
|
1093 | 1093 | # |
|
|
1094 | 1094 | # #print(mode,stdv) |
|
|
1095 | 1095 | # #Removing echoes greater than mode + std_factor*stdv |
|
|
1096 | 1096 | # noval = (abs(func2clean - mode)>=(factor_stdv*stdv)).nonzero() |
|
|
1097 | 1097 | # #noval tiene los indices que se van a remover |
|
|
1098 | 1098 | # #print("Chan ",ii," novals: ",len(noval[0])) |
|
|
1099 | 1099 | # if len(noval[0]) > 0: #forma de array (N,) es igual a longitud (N) |
|
|
1100 | 1100 | # novall = ((func2clean - mode) >= (factor_stdv*stdv)).nonzero() |
|
|
1101 | 1101 | # #print(novall) |
|
|
1102 | 1102 | # #print(" ",self.pairsArray[ii]) |
|
|
1103 | 1103 | # #cross_pairs = self.pairsArray[ii] |
|
|
1104 | 1104 | # #Getting coherent echoes which are removed. |
|
|
1105 | 1105 | # # if len(novall[0]) > 0: |
|
|
1106 | 1106 | # # |
|
|
1107 | 1107 | # # val_spc[novall[0],cross_pairs[0],ifreq,ih] = 1 |
|
|
1108 | 1108 | # # val_spc[novall[0],cross_pairs[1],ifreq,ih] = 1 |
|
|
1109 | 1109 | # # val_cspc[novall[0],ii,ifreq,ih] = 1 |
|
|
1110 | 1110 | # #print("OUT NOVALL 1") |
|
|
1111 | 1111 | # try: |
|
|
1112 | 1112 | # pair = (self.channels[ii],self.channels[ii + 1]) |
|
|
1113 | 1113 | # except: |
|
|
1114 | 1114 | # pair = (99,99) |
|
|
1115 | 1115 | # #print("par ", pair) |
|
|
1116 | 1116 | # if ( pair in self.crosspairs): |
|
|
1117 | 1117 | # q = self.crosspairs.index(pair) |
|
|
1118 | 1118 | # #print("estΓ‘ aqui: ", q, (ii,ii + 1)) |
|
|
1119 | 1119 | # new_a = numpy.delete(cspectra[:,q,ifreq,ih], noval[0]) |
|
|
1120 | 1120 | # cspectra[noval,q,ifreq,ih] = numpy.mean(new_a) #mean CrossSpectra |
|
|
1121 | 1121 | # |
|
|
1122 | 1122 | # #if channelA_ok: |
|
|
1123 | 1123 | # #chA = self.channels.index(cross_pairs[0]) |
|
|
1124 | 1124 | # new_b = numpy.delete(spectra[:,ii,ifreq,ih], noval[0]) |
|
|
1125 | 1125 | # spectra[noval,ii,ifreq,ih] = numpy.mean(new_b) #mean Spectra Pair A |
|
|
1126 | 1126 | # #channelA_ok = False |
|
|
1127 | 1127 | # |
|
|
1128 | 1128 | # # chB = self.channels.index(cross_pairs[1]) |
|
|
1129 | 1129 | # # new_c = numpy.delete(spectra[:,chB,ifreq,ih], noval[0]) |
|
|
1130 | 1130 | # # spectra[noval,chB,ifreq,ih] = numpy.mean(new_c) #mean Spectra Pair B |
|
|
1131 | 1131 | # # |
|
|
1132 | 1132 | # # channelsProcssd.append(self.crosspairs[ii][0]) # save channel A |
|
|
1133 | 1133 | # # channelsProcssd.append(self.crosspairs[ii][1]) # save channel B |
|
|
1134 | 1134 | # ''' |
|
|
1135 | 1135 | # ###ONLY FOR TEST: |
|
|
1136 | 1136 | # if ifreq ==fft_rev and ih==hei_rev: #TO VIEW A SIGNLE FREQUENCY |
|
|
1137 | 1137 | # func2clean = 10*numpy.log10(numpy.absolute(spectra[:,ii,ifreq,ih])) |
|
|
1138 | 1138 | # y_dist,binstep = numpy.histogram(func2clean,bins=range(int(min_val),int(max_val+2),step)) |
|
|
1139 | 1139 | # axs2[row_ax,col_ax].plot(binstep[:-1],newY,color='red') |
|
|
1140 | 1140 | # axs2[row_ax,col_ax].plot(binstep[:-1],y_dist,color='green') |
|
|
1141 | 1141 | # axs2[row_ax,col_ax].set_title("CH "+str(self.channels[ii])) |
|
|
1142 | 1142 | # ''' |
|
|
1143 | 1143 | # ''' |
|
|
1144 | 1144 | # ###ONLY FOR TEST: |
|
|
1145 | 1145 | # col_ax += 1 #contador de ploteo columnas |
|
|
1146 | 1146 | # ##print(col_ax) |
|
|
1147 | 1147 | # ###ONLY FOR TEST: |
|
|
1148 | 1148 | # if ifreq ==fft_rev and ih==hei_rev: #TO VIEW A SIGNLE FREQUENCY |
|
|
1149 | 1149 | # title = str(dataOut.datatime)+" nFFT: "+str(ifreq)+" Alt: "+str(self.heights[ih])+ " km" |
|
|
1150 | 1150 | # title2 = str(dataOut.datatime)+" nFFT: "+str(ifreq)+" Alt: "+str(self.heights[ih])+ " km CLEANED" |
|
|
1151 | 1151 | # fig.suptitle(title) |
|
|
1152 | 1152 | # fig2.suptitle(title2) |
|
|
1153 | 1153 | # plt.show() |
|
|
1154 | 1154 | # ''' |
|
|
1155 | 1155 | # ################################################################################################## |
|
|
1156 | 1156 | # |
|
|
1157 | 1157 | # #print("Getting average of the spectra and cross-spectra from incoherent echoes.") |
|
|
1158 | 1158 | # out_spectra = numpy.zeros([self.nChan,self.nFFTPoints,self.nHeights], dtype=float) #+numpy.nan |
|
|
1159 | 1159 | # out_cspectra = numpy.zeros([self.nPairs,self.nFFTPoints,self.nHeights], dtype=complex) #+numpy.nan |
|
|
1160 | 1160 | # for ih in range(self.nHeights): |
|
|
1161 | 1161 | # for ifreq in range(self.nFFTPoints): |
|
|
1162 | 1162 | # for ich in range(self.nChan): |
|
|
1163 | 1163 | # tmp = spectra[:,ich,ifreq,ih] |
|
|
1164 | 1164 | # valid = (numpy.isfinite(tmp[:])==True).nonzero() |
|
|
1165 | 1165 | # |
|
|
1166 | 1166 | # if len(valid[0]) >0 : |
|
|
1167 | 1167 | # out_spectra[ich,ifreq,ih] = numpy.nansum(tmp)#/len(valid[0]) |
|
|
1168 | 1168 | # |
|
|
1169 | 1169 | # for icr in range(self.nPairs): |
|
|
1170 | 1170 | # tmp = numpy.squeeze(cspectra[:,icr,ifreq,ih]) |
|
|
1171 | 1171 | # valid = (numpy.isfinite(tmp)==True).nonzero() |
|
|
1172 | 1172 | # if len(valid[0]) > 0: |
|
|
1173 | 1173 | # out_cspectra[icr,ifreq,ih] = numpy.nansum(tmp)#/len(valid[0]) |
|
|
1174 | 1174 | # |
|
|
1175 | 1175 | # return out_spectra, out_cspectra |
|
|
1176 | 1176 | # |
|
|
1177 | 1177 | # def REM_ISOLATED_POINTS(self,array,rth): |
|
|
1178 | 1178 | # # import matplotlib.pyplot as plt |
|
|
1179 | 1179 | # if rth == None : |
|
|
1180 | 1180 | # rth = 4 |
|
|
1181 | 1181 | # #print("REM ISO") |
|
|
1182 | 1182 | # num_prof = len(array[0,:,0]) |
|
|
1183 | 1183 | # num_hei = len(array[0,0,:]) |
|
|
1184 | 1184 | # n2d = len(array[:,0,0]) |
|
|
1185 | 1185 | # |
|
|
1186 | 1186 | # for ii in range(n2d) : |
|
|
1187 | 1187 | # #print ii,n2d |
|
|
1188 | 1188 | # tmp = array[ii,:,:] |
|
|
1189 | 1189 | # #print tmp.shape, array[ii,101,:],array[ii,102,:] |
|
|
1190 | 1190 | # |
|
|
1191 | 1191 | # # fig = plt.figure(figsize=(6,5)) |
|
|
1192 | 1192 | # # left, bottom, width, height = 0.1, 0.1, 0.8, 0.8 |
|
|
1193 | 1193 | # # ax = fig.add_axes([left, bottom, width, height]) |
|
|
1194 | 1194 | # # x = range(num_prof) |
|
|
1195 | 1195 | # # y = range(num_hei) |
|
|
1196 | 1196 | # # cp = ax.contour(y,x,tmp) |
|
|
1197 | 1197 | # # ax.clabel(cp, inline=True,fontsize=10) |
|
|
1198 | 1198 | # # plt.show() |
|
|
1199 | 1199 | # |
|
|
1200 | 1200 | # #indxs = WHERE(FINITE(tmp) AND tmp GT 0,cindxs) |
|
|
1201 | 1201 | # tmp = numpy.reshape(tmp,num_prof*num_hei) |
|
|
1202 | 1202 | # indxs1 = (numpy.isfinite(tmp)==True).nonzero() |
|
|
1203 | 1203 | # indxs2 = (tmp > 0).nonzero() |
|
|
1204 | 1204 | # |
|
|
1205 | 1205 | # indxs1 = (indxs1[0]) |
|
|
1206 | 1206 | # indxs2 = indxs2[0] |
|
|
1207 | 1207 | # #indxs1 = numpy.array(indxs1[0]) |
|
|
1208 | 1208 | # #indxs2 = numpy.array(indxs2[0]) |
|
|
1209 | 1209 | # indxs = None |
|
|
1210 | 1210 | # #print indxs1 , indxs2 |
|
|
1211 | 1211 | # for iv in range(len(indxs2)): |
|
|
1212 | 1212 | # indv = numpy.array((indxs1 == indxs2[iv]).nonzero()) |
|
|
1213 | 1213 | # #print len(indxs2), indv |
|
|
1214 | 1214 | # if len(indv[0]) > 0 : |
|
|
1215 | 1215 | # indxs = numpy.concatenate((indxs,indxs2[iv]), axis=None) |
|
|
1216 | 1216 | # # print indxs |
|
|
1217 | 1217 | # indxs = indxs[1:] |
|
|
1218 | 1218 | # #print(indxs, len(indxs)) |
|
|
1219 | 1219 | # if len(indxs) < 4 : |
|
|
1220 | 1220 | # array[ii,:,:] = 0. |
|
|
1221 | 1221 | # return |
|
|
1222 | 1222 | # |
|
|
1223 | 1223 | # xpos = numpy.mod(indxs ,num_hei) |
|
|
1224 | 1224 | # ypos = (indxs / num_hei) |
|
|
1225 | 1225 | # sx = numpy.argsort(xpos) # Ordering respect to "x" (time) |
|
|
1226 | 1226 | # #print sx |
|
|
1227 | 1227 | # xpos = xpos[sx] |
|
|
1228 | 1228 | # ypos = ypos[sx] |
|
|
1229 | 1229 | # |
|
|
1230 | 1230 | # # *********************************** Cleaning isolated points ********************************** |
|
|
1231 | 1231 | # ic = 0 |
|
|
1232 | 1232 | # while True : |
|
|
1233 | 1233 | # r = numpy.sqrt(list(numpy.power((xpos[ic]-xpos),2)+ numpy.power((ypos[ic]-ypos),2))) |
|
|
1234 | 1234 | # #no_coh = WHERE(FINITE(r) AND (r LE rth),cno_coh) |
|
|
1235 | 1235 | # #plt.plot(r) |
|
|
1236 | 1236 | # #plt.show() |
|
|
1237 | 1237 | # no_coh1 = (numpy.isfinite(r)==True).nonzero() |
|
|
1238 | 1238 | # no_coh2 = (r <= rth).nonzero() |
|
|
1239 | 1239 | # #print r, no_coh1, no_coh2 |
|
|
1240 | 1240 | # no_coh1 = numpy.array(no_coh1[0]) |
|
|
1241 | 1241 | # no_coh2 = numpy.array(no_coh2[0]) |
|
|
1242 | 1242 | # no_coh = None |
|
|
1243 | 1243 | # #print valid1 , valid2 |
|
|
1244 | 1244 | # for iv in range(len(no_coh2)): |
|
|
1245 | 1245 | # indv = numpy.array((no_coh1 == no_coh2[iv]).nonzero()) |
|
|
1246 | 1246 | # if len(indv[0]) > 0 : |
|
|
1247 | 1247 | # no_coh = numpy.concatenate((no_coh,no_coh2[iv]), axis=None) |
|
|
1248 | 1248 | # no_coh = no_coh[1:] |
|
|
1249 | 1249 | # #print len(no_coh), no_coh |
|
|
1250 | 1250 | # if len(no_coh) < 4 : |
|
|
1251 | 1251 | # #print xpos[ic], ypos[ic], ic |
|
|
1252 | 1252 | # # plt.plot(r) |
|
|
1253 | 1253 | # # plt.show() |
|
|
1254 | 1254 | # xpos[ic] = numpy.nan |
|
|
1255 | 1255 | # ypos[ic] = numpy.nan |
|
|
1256 | 1256 | # |
|
|
1257 | 1257 | # ic = ic + 1 |
|
|
1258 | 1258 | # if (ic == len(indxs)) : |
|
|
1259 | 1259 | # break |
|
|
1260 | 1260 | # #print( xpos, ypos) |
|
|
1261 | 1261 | # |
|
|
1262 | 1262 | # indxs = (numpy.isfinite(list(xpos))==True).nonzero() |
|
|
1263 | 1263 | # #print indxs[0] |
|
|
1264 | 1264 | # if len(indxs[0]) < 4 : |
|
|
1265 | 1265 | # array[ii,:,:] = 0. |
|
|
1266 | 1266 | # return |
|
|
1267 | 1267 | # |
|
|
1268 | 1268 | # xpos = xpos[indxs[0]] |
|
|
1269 | 1269 | # ypos = ypos[indxs[0]] |
|
|
1270 | 1270 | # for i in range(0,len(ypos)): |
|
|
1271 | 1271 | # ypos[i]=int(ypos[i]) |
|
|
1272 | 1272 | # junk = tmp |
|
|
1273 | 1273 | # tmp = junk*0.0 |
|
|
1274 | 1274 | # |
|
|
1275 | 1275 | # tmp[list(xpos + (ypos*num_hei))] = junk[list(xpos + (ypos*num_hei))] |
|
|
1276 | 1276 | # array[ii,:,:] = numpy.reshape(tmp,(num_prof,num_hei)) |
|
|
1277 | 1277 | # |
|
|
1278 | 1278 | # #print array.shape |
|
|
1279 | 1279 | # #tmp = numpy.reshape(tmp,(num_prof,num_hei)) |
|
|
1280 | 1280 | # #print tmp.shape |
|
|
1281 | 1281 | # |
|
|
1282 | 1282 | # # fig = plt.figure(figsize=(6,5)) |
|
|
1283 | 1283 | # # left, bottom, width, height = 0.1, 0.1, 0.8, 0.8 |
|
|
1284 | 1284 | # # ax = fig.add_axes([left, bottom, width, height]) |
|
|
1285 | 1285 | # # x = range(num_prof) |
|
|
1286 | 1286 | # # y = range(num_hei) |
|
|
1287 | 1287 | # # cp = ax.contour(y,x,array[ii,:,:]) |
|
|
1288 | 1288 | # # ax.clabel(cp, inline=True,fontsize=10) |
|
|
1289 | 1289 | # # plt.show() |
|
|
1290 | 1290 | # return array |
|
|
1291 | 1291 | # |
|
|
1292 | 1292 | |
|
|
1293 | 1293 | class IntegrationFaradaySpectra(Operation): |
|
|
1294 | 1294 | |
|
|
1295 | 1295 | __profIndex = 0 |
|
|
1296 | 1296 | __withOverapping = False |
|
|
1297 | 1297 | |
|
|
1298 | 1298 | __byTime = False |
|
|
1299 | 1299 | __initime = None |
|
|
1300 | 1300 | __lastdatatime = None |
|
|
1301 | 1301 | __integrationtime = None |
|
|
1302 | 1302 | |
|
|
1303 | 1303 | __buffer_spc = None |
|
|
1304 | 1304 | __buffer_cspc = None |
|
|
1305 | 1305 | __buffer_dc = None |
|
|
1306 | 1306 | |
|
|
1307 | 1307 | __dataReady = False |
|
|
1308 | 1308 | |
|
|
1309 | 1309 | __timeInterval = None |
|
|
1310 | 1310 | n_ints = None #matriz de numero de integracions (CH,HEI) |
|
|
1311 | 1311 | n = None |
|
|
1312 | 1312 | minHei_ind = None |
|
|
1313 | 1313 | maxHei_ind = None |
|
|
1314 | 1314 | navg = 1.0 |
|
|
1315 | 1315 | factor = 0.0 |
|
|
1316 | 1316 | dataoutliers = None # (CHANNELS, HEIGHTS) |
|
|
1317 | 1317 | |
|
|
1318 | 1318 | _flagProfilesByRange = False |
|
|
1319 | 1319 | _nProfilesByRange = 0 |
|
|
1320 | 1320 | |
|
|
1321 | 1321 | def __init__(self): |
|
|
1322 | 1322 | |
|
|
1323 | 1323 | Operation.__init__(self) |
|
|
1324 | 1324 | |
|
|
1325 | 1325 | def setup(self, dataOut,n=None, timeInterval=None, overlapping=False, DPL=None, minHei=None, maxHei=None, avg=1,factor=0.75): |
|
|
1326 | 1326 | """ |
|
|
1327 | 1327 | Set the parameters of the integration class. |
|
|
1328 | 1328 | |
|
|
1329 | 1329 | Inputs: |
|
|
1330 | 1330 | |
|
|
1331 | 1331 | n : Number of coherent integrations |
|
|
1332 | 1332 | timeInterval : Time of integration. If the parameter "n" is selected this one does not work |
|
|
1333 | 1333 | overlapping : |
|
|
1334 | 1334 | |
|
|
1335 | 1335 | """ |
|
|
1336 | 1336 | |
|
|
1337 | 1337 | self.__initime = None |
|
|
1338 | 1338 | self.__lastdatatime = 0 |
|
|
1339 | 1339 | |
|
|
1340 | 1340 | self.__buffer_spc = [] |
|
|
1341 | 1341 | self.__buffer_cspc = [] |
|
|
1342 | 1342 | self.__buffer_dc = 0 |
|
|
1343 | 1343 | |
|
|
1344 | 1344 | self.__profIndex = 0 |
|
|
1345 | 1345 | self.__dataReady = False |
|
|
1346 | 1346 | self.__byTime = False |
|
|
1347 | 1347 | |
|
|
1348 | 1348 | self.factor = factor |
|
|
1349 | 1349 | self.navg = avg |
|
|
1350 | 1350 | #self.ByLags = dataOut.ByLags ###REDEFINIR |
|
|
1351 | 1351 | self.ByLags = False |
|
|
1352 | 1352 | self.maxProfilesInt = 0 |
|
|
1353 | 1353 | self.__nChannels = dataOut.nChannels |
|
|
1354 | 1354 | if DPL != None: |
|
|
1355 | 1355 | self.DPL=DPL |
|
|
1356 | 1356 | else: |
|
|
1357 | 1357 | #self.DPL=dataOut.DPL ###REDEFINIR |
|
|
1358 | 1358 | self.DPL=0 |
|
|
1359 | 1359 | |
|
|
1360 | 1360 | if n is None and timeInterval is None: |
|
|
1361 | 1361 | raise ValueError("n or timeInterval should be specified ...") |
|
|
1362 | 1362 | |
|
|
1363 | 1363 | if n is not None: |
|
|
1364 | 1364 | self.n = int(n) |
|
|
1365 | 1365 | else: |
|
|
1366 | 1366 | self.__integrationtime = int(timeInterval) |
|
|
1367 | 1367 | self.n = None |
|
|
1368 | 1368 | self.__byTime = True |
|
|
1369 | 1369 | |
|
|
1370 | 1370 | |
|
|
1371 | 1371 | if minHei == None: |
|
|
1372 | 1372 | minHei = self.dataOut.heightList[0] |
|
|
1373 | 1373 | |
|
|
1374 | 1374 | if maxHei == None: |
|
|
1375 | 1375 | maxHei = self.dataOut.heightList[-1] |
|
|
1376 | 1376 | |
|
|
1377 | 1377 | if (minHei < self.dataOut.heightList[0]) or (minHei > maxHei): |
|
|
1378 | 1378 | print('minHei: %.2f is out of the heights range' % (minHei)) |
|
|
1379 | 1379 | print('minHei is setting to %.2f' % (self.dataOut.heightList[0])) |
|
|
1380 | 1380 | minHei = self.dataOut.heightList[0] |
|
|
1381 | 1381 | |
|
|
1382 | 1382 | if (maxHei > self.dataOut.heightList[-1]) or (maxHei < minHei): |
|
|
1383 | 1383 | print('maxHei: %.2f is out of the heights range' % (maxHei)) |
|
|
1384 | 1384 | print('maxHei is setting to %.2f' % (self.dataOut.heightList[-1])) |
|
|
1385 | 1385 | maxHei = self.dataOut.heightList[-1] |
|
|
1386 | 1386 | |
|
|
1387 | 1387 | ind_list1 = numpy.where(self.dataOut.heightList >= minHei) |
|
|
1388 | 1388 | ind_list2 = numpy.where(self.dataOut.heightList <= maxHei) |
|
|
1389 | 1389 | self.minHei_ind = ind_list1[0][0] |
|
|
1390 | 1390 | self.maxHei_ind = ind_list2[0][-1] |
|
|
1391 | 1391 | |
|
|
1392 | 1392 | def putData(self, data_spc, data_cspc, data_dc): |
|
|
1393 | 1393 | """ |
|
|
1394 | 1394 | Add a profile to the __buffer_spc and increase in one the __profileIndex |
|
|
1395 | 1395 | |
|
|
1396 | 1396 | """ |
|
|
1397 | 1397 | |
|
|
1398 | 1398 | self.__buffer_spc.append(data_spc) |
|
|
1399 | 1399 | |
|
|
1400 | 1400 | if self.__nChannels < 2: |
|
|
1401 | 1401 | self.__buffer_cspc = None |
|
|
1402 | 1402 | else: |
|
|
1403 | 1403 | self.__buffer_cspc.append(data_cspc) |
|
|
1404 | 1404 | |
|
|
1405 | 1405 | if data_dc is None: |
|
|
1406 | 1406 | self.__buffer_dc = None |
|
|
1407 | 1407 | else: |
|
|
1408 | 1408 | self.__buffer_dc += data_dc |
|
|
1409 | 1409 | |
|
|
1410 | 1410 | self.__profIndex += 1 |
|
|
1411 | 1411 | |
|
|
1412 | 1412 | return |
|
|
1413 | 1413 | |
|
|
1414 | 1414 | def hildebrand_sekhon_Integration(self,sortdata,navg, factor): |
|
|
1415 | 1415 | #data debe estar ordenado |
|
|
1416 | 1416 | #sortdata = numpy.sort(data, axis=None) |
|
|
1417 | 1417 | #sortID=data.argsort() |
|
|
1418 | 1418 | lenOfData = len(sortdata) |
|
|
1419 | 1419 | nums_min = lenOfData*factor |
|
|
1420 | 1420 | if nums_min <= 5: |
|
|
1421 | 1421 | nums_min = 5 |
|
|
1422 | 1422 | sump = 0. |
|
|
1423 | 1423 | sumq = 0. |
|
|
1424 | 1424 | j = 0 |
|
|
1425 | 1425 | cont = 1 |
|
|
1426 | 1426 | while((cont == 1)and(j < lenOfData)): |
|
|
1427 | 1427 | sump += sortdata[j] |
|
|
1428 | 1428 | sumq += sortdata[j]**2 |
|
|
1429 | 1429 | if j > nums_min: |
|
|
1430 | 1430 | rtest = float(j)/(j-1) + 1.0/navg |
|
|
1431 | 1431 | if ((sumq*j) > (rtest*sump**2)): |
|
|
1432 | 1432 | j = j - 1 |
|
|
1433 | 1433 | sump = sump - sortdata[j] |
|
|
1434 | 1434 | sumq = sumq - sortdata[j]**2 |
|
|
1435 | 1435 | cont = 0 |
|
|
1436 | 1436 | j += 1 |
|
|
1437 | 1437 | #lnoise = sump / j |
|
|
1438 | 1438 | #print("H S done") |
|
|
1439 | 1439 | #return j,sortID |
|
|
1440 | 1440 | return j |
|
|
1441 | 1441 | |
|
|
1442 | 1442 | |
|
|
1443 | 1443 | def pushData(self): |
|
|
1444 | 1444 | """ |
|
|
1445 | 1445 | Return the sum of the last profiles and the profiles used in the sum. |
|
|
1446 | 1446 | |
|
|
1447 | 1447 | Affected: |
|
|
1448 | 1448 | |
|
|
1449 | 1449 | self.__profileIndex |
|
|
1450 | 1450 | |
|
|
1451 | 1451 | """ |
|
|
1452 | 1452 | bufferH=None |
|
|
1453 | 1453 | buffer=None |
|
|
1454 | 1454 | buffer1=None |
|
|
1455 | 1455 | buffer_cspc=None |
|
|
1456 | 1456 | #print("aes: ", self.__buffer_cspc) |
|
|
1457 | 1457 | self.__buffer_spc=numpy.array(self.__buffer_spc) |
|
|
1458 | 1458 | if self.__nChannels > 1 : |
|
|
1459 | 1459 | self.__buffer_cspc=numpy.array(self.__buffer_cspc) |
|
|
1460 | 1460 | |
|
|
1461 | 1461 | #print("FREQ_DC",self.__buffer_spc.shape,self.__buffer_cspc.shape) |
|
|
1462 | 1462 | |
|
|
1463 | 1463 | freq_dc = int(self.__buffer_spc.shape[2] / 2) |
|
|
1464 | 1464 | #print("FREQ_DC",freq_dc,self.__buffer_spc.shape,self.nHeights) |
|
|
1465 | 1465 | |
|
|
1466 | 1466 | self.dataOutliers = numpy.zeros((self.nChannels,self.nHeights)) # --> almacen de outliers |
|
|
1467 | 1467 | |
|
|
1468 | 1468 | for k in range(self.minHei_ind,self.maxHei_ind): |
|
|
1469 | 1469 | if self.__nChannels > 1: |
|
|
1470 | 1470 | buffer_cspc=numpy.copy(self.__buffer_cspc[:,:,:,k]) |
|
|
1471 | 1471 | |
|
|
1472 | 1472 | outliers_IDs_cspc=[] |
|
|
1473 | 1473 | cspc_outliers_exist=False |
|
|
1474 | 1474 | for i in range(self.nChannels):#dataOut.nChannels): |
|
|
1475 | 1475 | |
|
|
1476 | 1476 | buffer1=numpy.copy(self.__buffer_spc[:,i,:,k]) |
|
|
1477 | 1477 | indexes=[] |
|
|
1478 | 1478 | #sortIDs=[] |
|
|
1479 | 1479 | outliers_IDs=[] |
|
|
1480 | 1480 | |
|
|
1481 | 1481 | for j in range(self.nProfiles): #frecuencias en el tiempo |
|
|
1482 | 1482 | # if i==0 and j==freq_dc: #NOT CONSIDERING DC PROFILE AT CHANNEL 0 |
|
|
1483 | 1483 | # continue |
|
|
1484 | 1484 | # if i==1 and j==0: #NOT CONSIDERING DC PROFILE AT CHANNEL 1 |
|
|
1485 | 1485 | # continue |
|
|
1486 | 1486 | buffer=buffer1[:,j] |
|
|
1487 | 1487 | sortdata = numpy.sort(buffer, axis=None) |
|
|
1488 | 1488 | |
|
|
1489 | 1489 | sortID=buffer.argsort() |
|
|
1490 | 1490 | index = _noise.hildebrand_sekhon2(sortdata,self.navg) |
|
|
1491 | 1491 | |
|
|
1492 | 1492 | #index,sortID=self.hildebrand_sekhon_Integration(buffer,1,self.factor) |
|
|
1493 | 1493 | |
|
|
1494 | 1494 | # fig,ax = plt.subplots() |
|
|
1495 | 1495 | # ax.set_title(str(k)+" "+str(j)) |
|
|
1496 | 1496 | # x=range(len(sortdata)) |
|
|
1497 | 1497 | # ax.scatter(x,sortdata) |
|
|
1498 | 1498 | # ax.axvline(index) |
|
|
1499 | 1499 | # plt.show() |
|
|
1500 | 1500 | |
|
|
1501 | 1501 | indexes.append(index) |
|
|
1502 | 1502 | #sortIDs.append(sortID) |
|
|
1503 | 1503 | outliers_IDs=numpy.append(outliers_IDs,sortID[index:]) |
|
|
1504 | 1504 | |
|
|
1505 | 1505 | #print("Outliers: ",outliers_IDs) |
|
|
1506 | 1506 | outliers_IDs=numpy.array(outliers_IDs) |
|
|
1507 | 1507 | outliers_IDs=outliers_IDs.ravel() |
|
|
1508 | 1508 | outliers_IDs=numpy.unique(outliers_IDs) |
|
|
1509 | 1509 | outliers_IDs=outliers_IDs.astype(numpy.dtype('int64')) |
|
|
1510 | 1510 | indexes=numpy.array(indexes) |
|
|
1511 | 1511 | indexmin=numpy.min(indexes) |
|
|
1512 | 1512 | |
|
|
1513 | 1513 | |
|
|
1514 | 1514 | #print(indexmin,buffer1.shape[0], k) |
|
|
1515 | 1515 | |
|
|
1516 | 1516 | # fig,ax = plt.subplots() |
|
|
1517 | 1517 | # ax.plot(sortdata) |
|
|
1518 | 1518 | # ax2 = ax.twinx() |
|
|
1519 | 1519 | # x=range(len(indexes)) |
|
|
1520 | 1520 | # #plt.scatter(x,indexes) |
|
|
1521 | 1521 | # ax2.scatter(x,indexes) |
|
|
1522 | 1522 | # plt.show() |
|
|
1523 | 1523 | |
|
|
1524 | 1524 | if indexmin != buffer1.shape[0]: |
|
|
1525 | 1525 | if self.__nChannels > 1: |
|
|
1526 | 1526 | cspc_outliers_exist= True |
|
|
1527 | 1527 | |
|
|
1528 | 1528 | lt=outliers_IDs |
|
|
1529 | 1529 | #avg=numpy.mean(buffer1[[t for t in range(buffer1.shape[0]) if t not in lt],:],axis=0) |
|
|
1530 | 1530 | |
|
|
1531 | 1531 | for p in list(outliers_IDs): |
|
|
1532 | 1532 | #buffer1[p,:]=avg |
|
|
1533 | 1533 | buffer1[p,:] = numpy.NaN |
|
|
1534 | 1534 | |
|
|
1535 | 1535 | self.dataOutliers[i,k] = len(outliers_IDs) |
|
|
1536 | 1536 | |
|
|
1537 | 1537 | |
|
|
1538 | 1538 | self.__buffer_spc[:,i,:,k]=numpy.copy(buffer1) |
|
|
1539 | 1539 | |
|
|
1540 | 1540 | |
|
|
1541 | 1541 | if self.__nChannels > 1: |
|
|
1542 | 1542 | outliers_IDs_cspc=numpy.append(outliers_IDs_cspc,outliers_IDs) |
|
|
1543 | 1543 | |
|
|
1544 | 1544 | |
|
|
1545 | 1545 | if self.__nChannels > 1: |
|
|
1546 | 1546 | outliers_IDs_cspc=outliers_IDs_cspc.astype(numpy.dtype('int64')) |
|
|
1547 | 1547 | if cspc_outliers_exist: |
|
|
1548 | 1548 | |
|
|
1549 | 1549 | lt=outliers_IDs_cspc |
|
|
1550 | 1550 | |
|
|
1551 | 1551 | #avg=numpy.mean(buffer_cspc[[t for t in range(buffer_cspc.shape[0]) if t not in lt],:],axis=0) |
|
|
1552 | 1552 | for p in list(outliers_IDs_cspc): |
|
|
1553 | 1553 | #buffer_cspc[p,:]=avg |
|
|
1554 | 1554 | buffer_cspc[p,:] = numpy.NaN |
|
|
1555 | 1555 | |
|
|
1556 | 1556 | if self.__nChannels > 1: |
|
|
1557 | 1557 | self.__buffer_cspc[:,:,:,k]=numpy.copy(buffer_cspc) |
|
|
1558 | 1558 | |
|
|
1559 | 1559 | |
|
|
1560 | 1560 | |
|
|
1561 | 1561 | |
|
|
1562 | 1562 | nOutliers = len(outliers_IDs) |
|
|
1563 | 1563 | #print("Outliers n: ",self.dataOutliers,nOutliers) |
|
|
1564 | 1564 | buffer=None |
|
|
1565 | 1565 | bufferH=None |
|
|
1566 | 1566 | buffer1=None |
|
|
1567 | 1567 | buffer_cspc=None |
|
|
1568 | 1568 | |
|
|
1569 | 1569 | |
|
|
1570 | 1570 | buffer=None |
|
|
1571 | 1571 | |
|
|
1572 | 1572 | #data_spc = numpy.sum(self.__buffer_spc,axis=0) |
|
|
1573 | 1573 | data_spc = numpy.nansum(self.__buffer_spc,axis=0) |
|
|
1574 | 1574 | if self.__nChannels > 1: |
|
|
1575 | 1575 | #data_cspc = numpy.sum(self.__buffer_cspc,axis=0) |
|
|
1576 | 1576 | data_cspc = numpy.nansum(self.__buffer_cspc,axis=0) |
|
|
1577 | 1577 | else: |
|
|
1578 | 1578 | data_cspc = None |
|
|
1579 | 1579 | data_dc = self.__buffer_dc |
|
|
1580 | 1580 | #(CH, HEIGH) |
|
|
1581 | 1581 | self.maxProfilesInt = self.__profIndex - 1 |
|
|
1582 | 1582 | n = self.__profIndex - self.dataOutliers # n becomes a matrix |
|
|
1583 | 1583 | |
|
|
1584 | 1584 | self.__buffer_spc = [] |
|
|
1585 | 1585 | self.__buffer_cspc = [] |
|
|
1586 | 1586 | self.__buffer_dc = 0 |
|
|
1587 | 1587 | self.__profIndex = 0 |
|
|
1588 | 1588 | #print("cleaned ",data_cspc) |
|
|
1589 | 1589 | return data_spc, data_cspc, data_dc, n |
|
|
1590 | 1590 | |
|
|
1591 | 1591 | def byProfiles(self, *args): |
|
|
1592 | 1592 | |
|
|
1593 | 1593 | self.__dataReady = False |
|
|
1594 | 1594 | avgdata_spc = None |
|
|
1595 | 1595 | avgdata_cspc = None |
|
|
1596 | 1596 | avgdata_dc = None |
|
|
1597 | 1597 | |
|
|
1598 | 1598 | self.putData(*args) |
|
|
1599 | 1599 | |
|
|
1600 | 1600 | if self.__profIndex >= self.n: |
|
|
1601 | 1601 | |
|
|
1602 | 1602 | avgdata_spc, avgdata_cspc, avgdata_dc, n = self.pushData() |
|
|
1603 | 1603 | self.n_ints = n |
|
|
1604 | 1604 | self.__dataReady = True |
|
|
1605 | 1605 | |
|
|
1606 | 1606 | return avgdata_spc, avgdata_cspc, avgdata_dc |
|
|
1607 | 1607 | |
|
|
1608 | 1608 | def byTime(self, datatime, *args): |
|
|
1609 | 1609 | |
|
|
1610 | 1610 | self.__dataReady = False |
|
|
1611 | 1611 | avgdata_spc = None |
|
|
1612 | 1612 | avgdata_cspc = None |
|
|
1613 | 1613 | avgdata_dc = None |
|
|
1614 | 1614 | |
|
|
1615 | 1615 | self.putData(*args) |
|
|
1616 | 1616 | |
|
|
1617 | 1617 | if (datatime - self.__initime) >= self.__integrationtime: |
|
|
1618 | 1618 | avgdata_spc, avgdata_cspc, avgdata_dc, n = self.pushData() |
|
|
1619 | 1619 | self.n_ints = n |
|
|
1620 | 1620 | self.__dataReady = True |
|
|
1621 | 1621 | |
|
|
1622 | 1622 | return avgdata_spc, avgdata_cspc, avgdata_dc |
|
|
1623 | 1623 | |
|
|
1624 | 1624 | def integrate(self, datatime, *args): |
|
|
1625 | 1625 | |
|
|
1626 | 1626 | if self.__profIndex == 0: |
|
|
1627 | 1627 | self.__initime = datatime |
|
|
1628 | 1628 | |
|
|
1629 | 1629 | if self.__byTime: |
|
|
1630 | 1630 | avgdata_spc, avgdata_cspc, avgdata_dc = self.byTime( |
|
|
1631 | 1631 | datatime, *args) |
|
|
1632 | 1632 | else: |
|
|
1633 | 1633 | avgdata_spc, avgdata_cspc, avgdata_dc = self.byProfiles(*args) |
|
|
1634 | 1634 | |
|
|
1635 | 1635 | if not self.__dataReady: |
|
|
1636 | 1636 | return None, None, None, None |
|
|
1637 | 1637 | |
|
|
1638 | 1638 | #print("integrate", avgdata_cspc) |
|
|
1639 | 1639 | return self.__initime, avgdata_spc, avgdata_cspc, avgdata_dc |
|
|
1640 | 1640 | |
|
|
1641 | 1641 | def run(self, dataOut, n=None, DPL = None,timeInterval=None, overlapping=False, minHei=None, maxHei=None, avg=1, factor=0.75): |
|
|
1642 | 1642 | self.dataOut = dataOut |
|
|
1643 | 1643 | if n == 1: |
|
|
1644 | 1644 | return self.dataOut |
|
|
1645 | 1645 | self.dataOut.processingHeaderObj.timeIncohInt = timeInterval |
|
|
1646 | 1646 | |
|
|
1647 | 1647 | if dataOut.flagProfilesByRange: |
|
|
1648 | 1648 | self._flagProfilesByRange = True |
|
|
1649 | 1649 | |
|
|
1650 | 1650 | if self.dataOut.nChannels == 1: |
|
|
1651 | 1651 | self.dataOut.data_cspc = None #si es un solo canal no vale la pena acumular DATOS |
|
|
1652 | 1652 | #print("IN spc:", self.dataOut.data_spc.shape, self.dataOut.data_cspc) |
|
|
1653 | 1653 | if not self.isConfig: |
|
|
1654 | 1654 | self.setup(self.dataOut, n, timeInterval, overlapping,DPL ,minHei, maxHei, avg, factor) |
|
|
1655 | 1655 | self.isConfig = True |
|
|
1656 | 1656 | |
|
|
1657 | 1657 | if not self.ByLags: |
|
|
1658 | 1658 | self.nProfiles=self.dataOut.nProfiles |
|
|
1659 | 1659 | self.nChannels=self.dataOut.nChannels |
|
|
1660 | 1660 | self.nHeights=self.dataOut.nHeights |
|
|
1661 | 1661 | avgdatatime, avgdata_spc, avgdata_cspc, avgdata_dc = self.integrate(self.dataOut.utctime, |
|
|
1662 | 1662 | self.dataOut.data_spc, |
|
|
1663 | 1663 | self.dataOut.data_cspc, |
|
|
1664 | 1664 | self.dataOut.data_dc) |
|
|
1665 | 1665 | else: |
|
|
1666 | 1666 | self.nProfiles=self.dataOut.nProfiles |
|
|
1667 | 1667 | self.nChannels=self.dataOut.nChannels |
|
|
1668 | 1668 | self.nHeights=self.dataOut.nHeights |
|
|
1669 | 1669 | avgdatatime, avgdata_spc, avgdata_cspc, avgdata_dc = self.integrate(self.dataOut.utctime, |
|
|
1670 | 1670 | self.dataOut.dataLag_spc, |
|
|
1671 | 1671 | self.dataOut.dataLag_cspc, |
|
|
1672 | 1672 | self.dataOut.dataLag_dc) |
|
|
1673 | 1673 | self.dataOut.flagNoData = True |
|
|
1674 | 1674 | |
|
|
1675 | 1675 | if self._flagProfilesByRange: |
|
|
1676 | 1676 | dataOut.flagProfilesByRange = True |
|
|
1677 | 1677 | self._nProfilesByRange += dataOut.nProfilesByRange |
|
|
1678 | 1678 | |
|
|
1679 | 1679 | if self.__dataReady: |
|
|
1680 | 1680 | |
|
|
1681 | 1681 | if not self.ByLags: |
|
|
1682 | 1682 | if self.nChannels == 1: |
|
|
1683 | 1683 | #print("f int", avgdata_spc.shape) |
|
|
1684 | 1684 | self.dataOut.data_spc = avgdata_spc |
|
|
1685 | 1685 | self.dataOut.data_cspc = None |
|
|
1686 | 1686 | else: |
|
|
1687 | 1687 | self.dataOut.data_spc = numpy.squeeze(avgdata_spc) |
|
|
1688 | 1688 | self.dataOut.data_cspc = numpy.squeeze(avgdata_cspc) |
|
|
1689 | 1689 | self.dataOut.data_dc = avgdata_dc |
|
|
1690 | 1690 | self.dataOut.data_outlier = self.dataOutliers |
|
|
1691 | 1691 | |
|
|
1692 | 1692 | |
|
|
1693 | 1693 | else: |
|
|
1694 | 1694 | self.dataOut.dataLag_spc = avgdata_spc |
|
|
1695 | 1695 | self.dataOut.dataLag_cspc = avgdata_cspc |
|
|
1696 | 1696 | self.dataOut.dataLag_dc = avgdata_dc |
|
|
1697 | 1697 | |
|
|
1698 | 1698 | self.dataOut.data_spc=self.dataOut.dataLag_spc[:,:,:,self.dataOut.LagPlot] |
|
|
1699 | 1699 | self.dataOut.data_cspc=self.dataOut.dataLag_cspc[:,:,:,self.dataOut.LagPlot] |
|
|
1700 | 1700 | self.dataOut.data_dc=self.dataOut.dataLag_dc[:,:,self.dataOut.LagPlot] |
|
|
1701 | 1701 | |
|
|
1702 | 1702 | self.dataOut.nIncohInt *= self.n_ints |
|
|
1703 | 1703 | #print("maxProfilesInt: ",self.maxProfilesInt) |
|
|
1704 | 1704 | |
|
|
1705 | 1705 | self.dataOut.utctime = avgdatatime |
|
|
1706 | 1706 | self.dataOut.flagNoData = False |
|
|
1707 | 1707 | |
|
|
1708 | 1708 | dataOut.nProfilesByRange = self._nProfilesByRange |
|
|
1709 | 1709 | self._nProfilesByRange = numpy.zeros((1,len(dataOut.heightList))) |
|
|
1710 | 1710 | self._flagProfilesByRange = False |
|
|
1711 | 1711 | |
|
|
1712 | 1712 | # #update Processing Header: |
|
|
1713 | 1713 | # self.dataOut.processingHeaderObj.nIncohInt = |
|
|
1714 | 1714 | # self.dataOut.processingHeaderObj.nFFTPoints = self.dataOut.nFFTPoints |
|
|
1715 | 1715 | |
|
|
1716 | 1716 | #print("Faraday Integration DONE...", self.dataOut.data_cspc) |
|
|
1717 | 1717 | #print(self.dataOut.flagNoData) |
|
|
1718 | 1718 | return self.dataOut |
|
|
1719 | 1719 | |
|
|
1720 | 1720 | |
|
|
1721 | 1721 | |
|
|
1722 | 1722 | class removeInterference(Operation): |
|
|
1723 | 1723 | |
|
|
1724 | 1724 | def removeInterference3(self, min_hei = None, max_hei = None): |
|
|
1725 | 1725 | |
|
|
1726 | 1726 | jspectra = self.dataOut.data_spc |
|
|
1727 | 1727 | #jcspectra = self.dataOut.data_cspc |
|
|
1728 | 1728 | jnoise = self.dataOut.getNoise() |
|
|
1729 | 1729 | num_incoh = self.dataOut.max_nIncohInt |
|
|
1730 | 1730 | #print(jspectra.shape) |
|
|
1731 | 1731 | num_channel, num_prof, num_hei = jspectra.shape |
|
|
1732 | 1732 | minHei = min_hei |
|
|
1733 | 1733 | maxHei = max_hei |
|
|
1734 | 1734 | ######################################################################## |
|
|
1735 | 1735 | if minHei == None or (minHei < self.dataOut.heightList[0]): |
|
|
1736 | 1736 | minHei = self.dataOut.heightList[0] |
|
|
1737 | 1737 | |
|
|
1738 | 1738 | if maxHei == None or (maxHei > self.dataOut.heightList[-1]): |
|
|
1739 | 1739 | maxHei = self.dataOut.heightList[-1] |
|
|
1740 | 1740 | minIndex = 0 |
|
|
1741 | 1741 | maxIndex = 0 |
|
|
1742 | 1742 | heights = self.dataOut.heightList |
|
|
1743 | 1743 | |
|
|
1744 | 1744 | inda = numpy.where(heights >= minHei) |
|
|
1745 | 1745 | indb = numpy.where(heights <= maxHei) |
|
|
1746 | 1746 | |
|
|
1747 | 1747 | try: |
|
|
1748 | 1748 | minIndex = inda[0][0] |
|
|
1749 | 1749 | except: |
|
|
1750 | 1750 | minIndex = 0 |
|
|
1751 | 1751 | try: |
|
|
1752 | 1752 | maxIndex = indb[0][-1] |
|
|
1753 | 1753 | except: |
|
|
1754 | 1754 | maxIndex = len(heights) |
|
|
1755 | 1755 | |
|
|
1756 | 1756 | if (minIndex < 0) or (minIndex > maxIndex): |
|
|
1757 | 1757 | raise ValueError("some value in (%d,%d) is not valid" % ( |
|
|
1758 | 1758 | minIndex, maxIndex)) |
|
|
1759 | 1759 | if (maxIndex >= self.dataOut.nHeights): |
|
|
1760 | 1760 | maxIndex = self.dataOut.nHeights - 1 |
|
|
1761 | 1761 | |
|
|
1762 | 1762 | ######################################################################## |
|
|
1763 | 1763 | |
|
|
1764 | 1764 | |
|
|
1765 | 1765 | #dataOut.max_nIncohInt * dataOut.nCohInt |
|
|
1766 | 1766 | if hasattr(self.dataOut.nIncohInt, 'shape'): |
|
|
1767 | 1767 | norm = self.dataOut.nIncohInt.T /self.dataOut.max_nIncohInt |
|
|
1768 | 1768 | norm = norm.T |
|
|
1769 | 1769 | else: |
|
|
1770 | 1770 | norm = self.dataOut.nIncohInt /self.dataOut.max_nIncohInt |
|
|
1771 | 1771 | norm = norm |
|
|
1772 | 1772 | |
|
|
1773 | 1773 | # Subrutina de Remocion de la Interferencia |
|
|
1774 | 1774 | for ich in range(num_channel): |
|
|
1775 | 1775 | # Se ordena los espectros segun su potencia (menor a mayor) |
|
|
1776 | 1776 | #power = jspectra[ich, mask_prof, :] |
|
|
1777 | 1777 | if hasattr(self.dataOut.nIncohInt, 'shape'): |
|
|
1778 | 1778 | interf = jspectra[ich, :, minIndex:maxIndex]/norm[ich,minIndex:maxIndex] |
|
|
1779 | 1779 | else: |
|
|
1780 | 1780 | interf = jspectra[ich, :, minIndex:maxIndex]/norm |
|
|
1781 | 1781 | # print(interf.shape) |
|
|
1782 | 1782 | inttef = interf.mean(axis=1) |
|
|
1783 | 1783 | |
|
|
1784 | 1784 | for hei in range(num_hei): |
|
|
1785 | 1785 | temp = jspectra[ich,:, hei]#/norm[ich,hei] |
|
|
1786 | 1786 | temp -= inttef |
|
|
1787 | 1787 | temp += jnoise[ich] |
|
|
1788 | 1788 | # print(jspectra.shape, temp.shape) |
|
|
1789 | 1789 | jspectra[ich,:, hei] = temp |
|
|
1790 | 1790 | |
|
|
1791 | 1791 | # Guardar Resultados |
|
|
1792 | 1792 | self.dataOut.data_spc = jspectra |
|
|
1793 | 1793 | #self.dataOut.data_cspc = jcspectra |
|
|
1794 | 1794 | |
|
|
1795 | 1795 | return 1 |
|
|
1796 | 1796 | |
|
|
1797 | 1797 | def removeInterference2(self): |
|
|
1798 | 1798 | |
|
|
1799 | 1799 | cspc = self.dataOut.data_cspc |
|
|
1800 | 1800 | spc = self.dataOut.data_spc |
|
|
1801 | 1801 | Heights = numpy.arange(cspc.shape[2]) |
|
|
1802 | 1802 | realCspc = numpy.abs(cspc) |
|
|
1803 | 1803 | |
|
|
1804 | 1804 | for i in range(cspc.shape[0]): |
|
|
1805 | 1805 | LinePower= numpy.sum(realCspc[i], axis=0) |
|
|
1806 | 1806 | Threshold = numpy.amax(LinePower)-numpy.sort(LinePower)[len(Heights)-int(len(Heights)*0.1)] |
|
|
1807 | 1807 | SelectedHeights = Heights[ numpy.where( LinePower < Threshold ) ] |
|
|
1808 | 1808 | InterferenceSum = numpy.sum( realCspc[i,:,SelectedHeights], axis=0 ) |
|
|
1809 | 1809 | InterferenceThresholdMin = numpy.sort(InterferenceSum)[int(len(InterferenceSum)*0.98)] |
|
|
1810 | 1810 | InterferenceThresholdMax = numpy.sort(InterferenceSum)[int(len(InterferenceSum)*0.99)] |
|
|
1811 | 1811 | |
|
|
1812 | 1812 | |
|
|
1813 | 1813 | InterferenceRange = numpy.where( ([InterferenceSum > InterferenceThresholdMin]))# , InterferenceSum < InterferenceThresholdMax]) ) |
|
|
1814 | 1814 | #InterferenceRange = numpy.where( ([InterferenceRange < InterferenceThresholdMax])) |
|
|
1815 | 1815 | if len(InterferenceRange)<int(cspc.shape[1]*0.3): |
|
|
1816 | 1816 | cspc[i,InterferenceRange,:] = numpy.NaN |
|
|
1817 | 1817 | |
|
|
1818 | 1818 | self.dataOut.data_cspc = cspc |
|
|
1819 | 1819 | |
|
|
1820 | 1820 | def removeInterference(self, interf = 2, hei_interf = None, nhei_interf = None, offhei_interf = None): |
|
|
1821 | 1821 | |
|
|
1822 | 1822 | jspectra = self.dataOut.data_spc |
|
|
1823 | 1823 | jcspectra = self.dataOut.data_cspc |
|
|
1824 | 1824 | jnoise = self.dataOut.getNoise() |
|
|
1825 | 1825 | #num_incoh = self.dataOut.nIncohInt |
|
|
1826 | 1826 | num_incoh = self.dataOut.max_nIncohInt |
|
|
1827 | 1827 | #print("spc: ", jspectra.shape, jcspectra) |
|
|
1828 | 1828 | num_channel = jspectra.shape[0] |
|
|
1829 | 1829 | num_prof = jspectra.shape[1] |
|
|
1830 | 1830 | num_hei = jspectra.shape[2] |
|
|
1831 | 1831 | |
|
|
1832 | 1832 | count_hei = nhei_interf |
|
|
1833 | 1833 | # hei_interf |
|
|
1834 | 1834 | if hei_interf is None: |
|
|
1835 | 1835 | count_hei = int(num_hei / 2) # a half of total ranges |
|
|
1836 | 1836 | hei_interf = numpy.asmatrix(list(range(count_hei))) + num_hei - count_hei |
|
|
1837 | 1837 | hei_interf = numpy.asarray(hei_interf)[0] |
|
|
1838 | 1838 | #print(hei_interf) |
|
|
1839 | 1839 | # nhei_interf |
|
|
1840 | 1840 | if (nhei_interf == None): |
|
|
1841 | 1841 | nhei_interf = 5 |
|
|
1842 | 1842 | if (nhei_interf < 1): |
|
|
1843 | 1843 | nhei_interf = 1 |
|
|
1844 | 1844 | if (nhei_interf > count_hei): |
|
|
1845 | 1845 | nhei_interf = count_hei |
|
|
1846 | 1846 | if (offhei_interf == None): |
|
|
1847 | 1847 | offhei_interf = 0 |
|
|
1848 | 1848 | |
|
|
1849 | 1849 | ind_hei = list(range(num_hei)) |
|
|
1850 | 1850 | # mask_prof = numpy.asarray(range(num_prof - 2)) + 1 |
|
|
1851 | 1851 | # mask_prof[range(num_prof/2 - 1,len(mask_prof))] += 1 |
|
|
1852 | 1852 | mask_prof = numpy.asarray(list(range(num_prof))) |
|
|
1853 | 1853 | num_mask_prof = mask_prof.size |
|
|
1854 | 1854 | comp_mask_prof = [0, num_prof / 2] |
|
|
1855 | 1855 | |
|
|
1856 | 1856 | # noise_exist: Determina si la variable jnoise ha sido definida y contiene la informacion del ruido de cada canal |
|
|
1857 | 1857 | if (jnoise.size < num_channel or numpy.isnan(jnoise).any()): |
|
|
1858 | 1858 | jnoise = numpy.nan |
|
|
1859 | 1859 | noise_exist = jnoise[0] < numpy.Inf |
|
|
1860 | 1860 | |
|
|
1861 | 1861 | # Subrutina de Remocion de la Interferencia |
|
|
1862 | 1862 | for ich in range(num_channel): |
|
|
1863 | 1863 | # Se ordena los espectros segun su potencia (menor a mayor) |
|
|
1864 | 1864 | power = jspectra[ich, mask_prof, :] |
|
|
1865 | 1865 | power = power[:, hei_interf] |
|
|
1866 | 1866 | power = power.sum(axis=0) |
|
|
1867 | 1867 | psort = power.ravel().argsort() |
|
|
1868 | 1868 | #print(hei_interf[psort[list(range(offhei_interf, nhei_interf + offhei_interf))]]) |
|
|
1869 | 1869 | # Se estima la interferencia promedio en los Espectros de Potencia empleando |
|
|
1870 | 1870 | junkspc_interf = jspectra[ich, :, hei_interf[psort[list(range( |
|
|
1871 | 1871 | offhei_interf, nhei_interf + offhei_interf))]]] |
|
|
1872 | 1872 | |
|
|
1873 | 1873 | if noise_exist: |
|
|
1874 | 1874 | # tmp_noise = jnoise[ich] / num_prof |
|
|
1875 | 1875 | tmp_noise = jnoise[ich] |
|
|
1876 | 1876 | junkspc_interf = junkspc_interf - tmp_noise |
|
|
1877 | 1877 | #junkspc_interf[:,comp_mask_prof] = 0 |
|
|
1878 | 1878 | #print(junkspc_interf.shape) |
|
|
1879 | 1879 | jspc_interf = junkspc_interf.sum(axis=0) / nhei_interf |
|
|
1880 | 1880 | jspc_interf = jspc_interf.transpose() |
|
|
1881 | 1881 | # Calculando el espectro de interferencia promedio |
|
|
1882 | 1882 | noiseid = numpy.where(jspc_interf <= tmp_noise / numpy.sqrt(num_incoh)) |
|
|
1883 | 1883 | noiseid = noiseid[0] |
|
|
1884 | 1884 | cnoiseid = noiseid.size |
|
|
1885 | 1885 | interfid = numpy.where(jspc_interf > tmp_noise / numpy.sqrt(num_incoh)) |
|
|
1886 | 1886 | interfid = interfid[0] |
|
|
1887 | 1887 | cinterfid = interfid.size |
|
|
1888 | 1888 | |
|
|
1889 | 1889 | if (cnoiseid > 0): |
|
|
1890 | 1890 | jspc_interf[noiseid] = 0 |
|
|
1891 | 1891 | # Expandiendo los perfiles a limpiar |
|
|
1892 | 1892 | if (cinterfid > 0): |
|
|
1893 | 1893 | new_interfid = ( |
|
|
1894 | 1894 | numpy.r_[interfid - 1, interfid, interfid + 1] + num_prof) % num_prof |
|
|
1895 | 1895 | new_interfid = numpy.asarray(new_interfid) |
|
|
1896 | 1896 | new_interfid = {x for x in new_interfid} |
|
|
1897 | 1897 | new_interfid = numpy.array(list(new_interfid)) |
|
|
1898 | 1898 | new_cinterfid = new_interfid.size |
|
|
1899 | 1899 | else: |
|
|
1900 | 1900 | new_cinterfid = 0 |
|
|
1901 | 1901 | |
|
|
1902 | 1902 | for ip in range(new_cinterfid): |
|
|
1903 | 1903 | ind = junkspc_interf[:, new_interfid[ip]].ravel().argsort() |
|
|
1904 | 1904 | jspc_interf[new_interfid[ip]] = junkspc_interf[ind[nhei_interf // 2], new_interfid[ip]] |
|
|
1905 | 1905 | |
|
|
1906 | 1906 | jspectra[ich, :, ind_hei] = jspectra[ich, :,ind_hei] - jspc_interf # Corregir indices |
|
|
1907 | 1907 | |
|
|
1908 | 1908 | # Removiendo la interferencia del punto de mayor interferencia |
|
|
1909 | 1909 | ListAux = jspc_interf[mask_prof].tolist() |
|
|
1910 | 1910 | maxid = ListAux.index(max(ListAux)) |
|
|
1911 | 1911 | #print(cinterfid) |
|
|
1912 | 1912 | if cinterfid > 0: |
|
|
1913 | 1913 | for ip in range(cinterfid * (interf == 2) - 1): |
|
|
1914 | 1914 | ind = (jspectra[ich, interfid[ip], :] < tmp_noise * |
|
|
1915 | 1915 | (1 + 1 / numpy.sqrt(num_incoh))).nonzero() |
|
|
1916 | 1916 | cind = len(ind) |
|
|
1917 | 1917 | |
|
|
1918 | 1918 | if (cind > 0): |
|
|
1919 | 1919 | jspectra[ich, interfid[ip], ind] = tmp_noise * \ |
|
|
1920 | 1920 | (1 + (numpy.random.uniform(cind) - 0.5) / |
|
|
1921 | 1921 | numpy.sqrt(num_incoh)) |
|
|
1922 | 1922 | |
|
|
1923 | 1923 | ind = numpy.array([-2, -1, 1, 2]) |
|
|
1924 | 1924 | xx = numpy.zeros([4, 4]) |
|
|
1925 | 1925 | |
|
|
1926 | 1926 | for id1 in range(4): |
|
|
1927 | 1927 | xx[:, id1] = ind[id1]**numpy.asarray(list(range(4))) |
|
|
1928 | 1928 | xx_inv = numpy.linalg.inv(xx) |
|
|
1929 | 1929 | xx = xx_inv[:, 0] |
|
|
1930 | 1930 | ind = (ind + maxid + num_mask_prof) % num_mask_prof |
|
|
1931 | 1931 | yy = jspectra[ich, mask_prof[ind], :] |
|
|
1932 | 1932 | jspectra[ich, mask_prof[maxid], :] = numpy.dot(yy.transpose(), xx) |
|
|
1933 | 1933 | |
|
|
1934 | 1934 | indAux = (jspectra[ich, :, :] < tmp_noise * |
|
|
1935 | 1935 | (1 - 1 / numpy.sqrt(num_incoh))).nonzero() |
|
|
1936 | 1936 | #print(indAux) |
|
|
1937 | 1937 | jspectra[ich, indAux[0], indAux[1]] = tmp_noise * \ |
|
|
1938 | 1938 | (1 - 1 / numpy.sqrt(num_incoh)) |
|
|
1939 | 1939 | |
|
|
1940 | 1940 | # Remocion de Interferencia en el Cross Spectra |
|
|
1941 | 1941 | if jcspectra is None: |
|
|
1942 | 1942 | return jspectra, jcspectra |
|
|
1943 | 1943 | num_pairs = int(jcspectra.size / (num_prof * num_hei)) |
|
|
1944 | 1944 | jcspectra = jcspectra.reshape(num_pairs, num_prof, num_hei) |
|
|
1945 | 1945 | |
|
|
1946 | 1946 | for ip in range(num_pairs): |
|
|
1947 | 1947 | |
|
|
1948 | 1948 | #------------------------------------------- |
|
|
1949 | 1949 | |
|
|
1950 | 1950 | cspower = numpy.abs(jcspectra[ip, mask_prof, :]) |
|
|
1951 | 1951 | cspower = cspower[:, hei_interf] |
|
|
1952 | 1952 | cspower = cspower.sum(axis=0) |
|
|
1953 | 1953 | |
|
|
1954 | 1954 | cspsort = cspower.ravel().argsort() |
|
|
1955 | 1955 | junkcspc_interf = jcspectra[ip, :, hei_interf[cspsort[list(range( |
|
|
1956 | 1956 | offhei_interf, nhei_interf + offhei_interf))]]] |
|
|
1957 | 1957 | junkcspc_interf = junkcspc_interf.transpose() |
|
|
1958 | 1958 | jcspc_interf = junkcspc_interf.sum(axis=1) / nhei_interf |
|
|
1959 | 1959 | |
|
|
1960 | 1960 | ind = numpy.abs(jcspc_interf[mask_prof]).ravel().argsort() |
|
|
1961 | 1961 | |
|
|
1962 | 1962 | median_real = int(numpy.median(numpy.real( |
|
|
1963 | 1963 | junkcspc_interf[mask_prof[ind[list(range(3 * num_prof // 4))]], :]))) |
|
|
1964 | 1964 | median_imag = int(numpy.median(numpy.imag( |
|
|
1965 | 1965 | junkcspc_interf[mask_prof[ind[list(range(3 * num_prof // 4))]], :]))) |
|
|
1966 | 1966 | comp_mask_prof = [int(e) for e in comp_mask_prof] |
|
|
1967 | 1967 | junkcspc_interf[comp_mask_prof, :] = numpy.complex( |
|
|
1968 | 1968 | median_real, median_imag) |
|
|
1969 | 1969 | |
|
|
1970 | 1970 | for iprof in range(num_prof): |
|
|
1971 | 1971 | ind = numpy.abs(junkcspc_interf[iprof, :]).ravel().argsort() |
|
|
1972 | 1972 | jcspc_interf[iprof] = junkcspc_interf[iprof, ind[nhei_interf // 2]] |
|
|
1973 | 1973 | |
|
|
1974 | 1974 | # Removiendo la Interferencia |
|
|
1975 | 1975 | jcspectra[ip, :, ind_hei] = jcspectra[ip, |
|
|
1976 | 1976 | :, ind_hei] - jcspc_interf |
|
|
1977 | 1977 | |
|
|
1978 | 1978 | ListAux = numpy.abs(jcspc_interf[mask_prof]).tolist() |
|
|
1979 | 1979 | maxid = ListAux.index(max(ListAux)) |
|
|
1980 | 1980 | |
|
|
1981 | 1981 | ind = numpy.array([-2, -1, 1, 2]) |
|
|
1982 | 1982 | xx = numpy.zeros([4, 4]) |
|
|
1983 | 1983 | |
|
|
1984 | 1984 | for id1 in range(4): |
|
|
1985 | 1985 | xx[:, id1] = ind[id1]**numpy.asarray(list(range(4))) |
|
|
1986 | 1986 | |
|
|
1987 | 1987 | xx_inv = numpy.linalg.inv(xx) |
|
|
1988 | 1988 | xx = xx_inv[:, 0] |
|
|
1989 | 1989 | |
|
|
1990 | 1990 | ind = (ind + maxid + num_mask_prof) % num_mask_prof |
|
|
1991 | 1991 | yy = jcspectra[ip, mask_prof[ind], :] |
|
|
1992 | 1992 | jcspectra[ip, mask_prof[maxid], :] = numpy.dot(yy.transpose(), xx) |
|
|
1993 | 1993 | |
|
|
1994 | 1994 | # Guardar Resultados |
|
|
1995 | 1995 | self.dataOut.data_spc = jspectra |
|
|
1996 | 1996 | self.dataOut.data_cspc = jcspectra |
|
|
1997 | 1997 | |
|
|
1998 | 1998 | return 1 |
|
|
1999 | 1999 | |
|
|
2000 | 2000 | def run(self, dataOut, interf = 2,hei_interf = None, nhei_interf = None, offhei_interf = None, mode=1, minHei=None, maxHei=None): |
|
|
2001 | 2001 | |
|
|
2002 | 2002 | self.dataOut = dataOut |
|
|
2003 | 2003 | |
|
|
2004 | 2004 | if mode == 1: |
|
|
2005 | 2005 | self.removeInterference(interf = 2,hei_interf = hei_interf, nhei_interf = nhei_interf, offhei_interf = offhei_interf) |
|
|
2006 | 2006 | elif mode == 2: |
|
|
2007 | 2007 | self.removeInterference2() |
|
|
2008 | 2008 | elif mode == 3: |
|
|
2009 | 2009 | self.removeInterference3(min_hei=minHei, max_hei=maxHei) |
|
|
2010 | 2010 | return self.dataOut |
|
|
2011 | 2011 | |
|
|
2012 | 2012 | |
|
|
2013 | 2013 | class IncohInt(Operation): |
|
|
2014 | 2014 | |
|
|
2015 | 2015 | __profIndex = 0 |
|
|
2016 | 2016 | __withOverapping = False |
|
|
2017 | 2017 | |
|
|
2018 | 2018 | __byTime = False |
|
|
2019 | 2019 | __initime = None |
|
|
2020 | 2020 | __lastdatatime = None |
|
|
2021 | 2021 | __integrationtime = None |
|
|
2022 | 2022 | |
|
|
2023 | 2023 | __buffer_spc = None |
|
|
2024 | 2024 | __buffer_cspc = None |
|
|
2025 | 2025 | __buffer_dc = None |
|
|
2026 | 2026 | |
|
|
2027 | 2027 | __dataReady = False |
|
|
2028 | 2028 | |
|
|
2029 | 2029 | __timeInterval = None |
|
|
2030 | 2030 | incohInt = 0 |
|
|
2031 | 2031 | nOutliers = 0 |
|
|
2032 | 2032 | n = None |
|
|
2033 | 2033 | |
|
|
2034 | 2034 | _flagProfilesByRange = False |
|
|
2035 | 2035 | _nProfilesByRange = 0 |
|
|
2036 | 2036 | def __init__(self): |
|
|
2037 | 2037 | |
|
|
2038 | 2038 | Operation.__init__(self) |
|
|
2039 | 2039 | |
|
|
2040 | 2040 | def setup(self, n=None, timeInterval=None, overlapping=False): |
|
|
2041 | 2041 | """ |
|
|
2042 | 2042 | Set the parameters of the integration class. |
|
|
2043 | 2043 | |
|
|
2044 | 2044 | Inputs: |
|
|
2045 | 2045 | |
|
|
2046 | 2046 | n : Number of coherent integrations |
|
|
2047 | 2047 | timeInterval : Time of integration. If the parameter "n" is selected this one does not work |
|
|
2048 | 2048 | overlapping : |
|
|
2049 | 2049 | |
|
|
2050 | 2050 | """ |
|
|
2051 | 2051 | |
|
|
2052 | 2052 | self.__initime = None |
|
|
2053 | 2053 | self.__lastdatatime = 0 |
|
|
2054 | 2054 | |
|
|
2055 | 2055 | self.__buffer_spc = 0 |
|
|
2056 | 2056 | self.__buffer_cspc = 0 |
|
|
2057 | 2057 | self.__buffer_dc = 0 |
|
|
2058 | 2058 | |
|
|
2059 | 2059 | self.__profIndex = 0 |
|
|
2060 | 2060 | self.__dataReady = False |
|
|
2061 | 2061 | self.__byTime = False |
|
|
2062 | 2062 | self.incohInt = 0 |
|
|
2063 | 2063 | self.nOutliers = 0 |
|
|
2064 | 2064 | if n is None and timeInterval is None: |
|
|
2065 | 2065 | raise ValueError("n or timeInterval should be specified ...") |
|
|
2066 | 2066 | |
|
|
2067 | 2067 | if n is not None: |
|
|
2068 | 2068 | self.n = int(n) |
|
|
2069 | 2069 | else: |
|
|
2070 | 2070 | |
|
|
2071 | 2071 | self.__integrationtime = int(timeInterval) |
|
|
2072 | 2072 | self.n = None |
|
|
2073 | 2073 | self.__byTime = True |
|
|
2074 | 2074 | |
|
|
2075 | 2075 | |
|
|
2076 | 2076 | |
|
|
2077 | 2077 | def putData(self, data_spc, data_cspc, data_dc): |
|
|
2078 | 2078 | """ |
|
|
2079 | 2079 | Add a profile to the __buffer_spc and increase in one the __profileIndex |
|
|
2080 | 2080 | |
|
|
2081 | 2081 | """ |
|
|
2082 | 2082 | if data_spc.all() == numpy.nan : |
|
|
2083 | 2083 | print("nan ") |
|
|
2084 | 2084 | return |
|
|
2085 | 2085 | self.__buffer_spc += data_spc |
|
|
2086 | 2086 | |
|
|
2087 | 2087 | if data_cspc is None: |
|
|
2088 | 2088 | self.__buffer_cspc = None |
|
|
2089 | 2089 | else: |
|
|
2090 | 2090 | self.__buffer_cspc += data_cspc |
|
|
2091 | 2091 | |
|
|
2092 | 2092 | if data_dc is None: |
|
|
2093 | 2093 | self.__buffer_dc = None |
|
|
2094 | 2094 | else: |
|
|
2095 | 2095 | self.__buffer_dc += data_dc |
|
|
2096 | 2096 | |
|
|
2097 | 2097 | self.__profIndex += 1 |
|
|
2098 | 2098 | |
|
|
2099 | 2099 | return |
|
|
2100 | 2100 | |
|
|
2101 | 2101 | def pushData(self): |
|
|
2102 | 2102 | """ |
|
|
2103 | 2103 | Return the sum of the last profiles and the profiles used in the sum. |
|
|
2104 | 2104 | |
|
|
2105 | 2105 | Affected: |
|
|
2106 | 2106 | |
|
|
2107 | 2107 | self.__profileIndex |
|
|
2108 | 2108 | |
|
|
2109 | 2109 | """ |
|
|
2110 | 2110 | |
|
|
2111 | 2111 | data_spc = self.__buffer_spc |
|
|
2112 | 2112 | data_cspc = self.__buffer_cspc |
|
|
2113 | 2113 | data_dc = self.__buffer_dc |
|
|
2114 | 2114 | n = self.__profIndex |
|
|
2115 | 2115 | |
|
|
2116 | 2116 | self.__buffer_spc = 0 |
|
|
2117 | 2117 | self.__buffer_cspc = 0 |
|
|
2118 | 2118 | self.__buffer_dc = 0 |
|
|
2119 | 2119 | |
|
|
2120 | 2120 | |
|
|
2121 | 2121 | return data_spc, data_cspc, data_dc, n |
|
|
2122 | 2122 | |
|
|
2123 | 2123 | def byProfiles(self, *args): |
|
|
2124 | 2124 | |
|
|
2125 | 2125 | self.__dataReady = False |
|
|
2126 | 2126 | avgdata_spc = None |
|
|
2127 | 2127 | avgdata_cspc = None |
|
|
2128 | 2128 | avgdata_dc = None |
|
|
2129 | 2129 | |
|
|
2130 | 2130 | self.putData(*args) |
|
|
2131 | 2131 | |
|
|
2132 | 2132 | if self.__profIndex == self.n: |
|
|
2133 | 2133 | |
|
|
2134 | 2134 | avgdata_spc, avgdata_cspc, avgdata_dc, n = self.pushData() |
|
|
2135 | 2135 | self.n = n |
|
|
2136 | 2136 | self.__dataReady = True |
|
|
2137 | 2137 | |
|
|
2138 | 2138 | return avgdata_spc, avgdata_cspc, avgdata_dc |
|
|
2139 | 2139 | |
|
|
2140 | 2140 | def byTime(self, datatime, *args): |
|
|
2141 | 2141 | |
|
|
2142 | 2142 | self.__dataReady = False |
|
|
2143 | 2143 | avgdata_spc = None |
|
|
2144 | 2144 | avgdata_cspc = None |
|
|
2145 | 2145 | avgdata_dc = None |
|
|
2146 | 2146 | |
|
|
2147 | 2147 | self.putData(*args) |
|
|
2148 | 2148 | |
|
|
2149 | 2149 | if (datatime - self.__initime) >= self.__integrationtime: |
|
|
2150 | 2150 | avgdata_spc, avgdata_cspc, avgdata_dc, n = self.pushData() |
|
|
2151 | 2151 | self.n = n |
|
|
2152 | 2152 | self.__dataReady = True |
|
|
2153 | 2153 | |
|
|
2154 | 2154 | return avgdata_spc, avgdata_cspc, avgdata_dc |
|
|
2155 | 2155 | |
|
|
2156 | 2156 | def integrate(self, datatime, *args): |
|
|
2157 | 2157 | |
|
|
2158 | 2158 | if self.__profIndex == 0: |
|
|
2159 | 2159 | self.__initime = datatime |
|
|
2160 | 2160 | |
|
|
2161 | 2161 | if self.__byTime: |
|
|
2162 | 2162 | avgdata_spc, avgdata_cspc, avgdata_dc = self.byTime( |
|
|
2163 | 2163 | datatime, *args) |
|
|
2164 | 2164 | else: |
|
|
2165 | 2165 | avgdata_spc, avgdata_cspc, avgdata_dc = self.byProfiles(*args) |
|
|
2166 | 2166 | |
|
|
2167 | 2167 | if not self.__dataReady: |
|
|
2168 | 2168 | return None, None, None, None |
|
|
2169 | 2169 | |
|
|
2170 | 2170 | return self.__initime, avgdata_spc, avgdata_cspc, avgdata_dc |
|
|
2171 | 2171 | |
|
|
2172 | 2172 | def run(self, dataOut, n=None, timeInterval=None, overlapping=False): |
|
|
2173 | 2173 | if n == 1: |
|
|
2174 | 2174 | return dataOut |
|
|
2175 | 2175 | |
|
|
2176 | 2176 | if dataOut.flagNoData == True: |
|
|
2177 | 2177 | return dataOut |
|
|
2178 | 2178 | |
|
|
2179 | 2179 | if dataOut.flagProfilesByRange == True: |
|
|
2180 | 2180 | self._flagProfilesByRange = True |
|
|
2181 | 2181 | |
|
|
2182 | 2182 | dataOut.flagNoData = True |
|
|
2183 | 2183 | dataOut.processingHeaderObj.timeIncohInt = timeInterval |
|
|
2184 | 2184 | if not self.isConfig: |
|
|
2185 | 2185 | self._nProfilesByRange = numpy.zeros((1,len(dataOut.heightList))) |
|
|
2186 | 2186 | self.setup(n, timeInterval, overlapping) |
|
|
2187 | 2187 | self.isConfig = True |
|
|
2188 | 2188 | |
|
|
2189 | 2189 | |
|
|
2190 | 2190 | avgdatatime, avgdata_spc, avgdata_cspc, avgdata_dc = self.integrate(dataOut.utctime, |
|
|
2191 | 2191 | dataOut.data_spc, |
|
|
2192 | 2192 | dataOut.data_cspc, |
|
|
2193 | 2193 | dataOut.data_dc) |
|
|
2194 | 2194 | |
|
|
2195 | 2195 | self.incohInt += dataOut.nIncohInt |
|
|
2196 | 2196 | |
|
|
2197 | 2197 | |
|
|
2198 | 2198 | if isinstance(dataOut.data_outlier,numpy.ndarray) or isinstance(dataOut.data_outlier,int) or isinstance(dataOut.data_outlier, float): |
|
|
2199 | 2199 | self.nOutliers += dataOut.data_outlier |
|
|
2200 | 2200 | |
|
|
2201 | 2201 | if self._flagProfilesByRange: |
|
|
2202 | 2202 | dataOut.flagProfilesByRange = True |
|
|
2203 | 2203 | self._nProfilesByRange += dataOut.nProfilesByRange |
|
|
2204 | 2204 | |
|
|
2205 | 2205 | if self.__dataReady: |
|
|
2206 | print("IncohInt Done ", self.incohInt) | |
|
|
2207 | 2206 | #print("prof: ",dataOut.max_nIncohInt,self.__profIndex) |
|
|
2208 | 2207 | dataOut.data_spc = avgdata_spc |
|
|
2209 | 2208 | dataOut.data_cspc = avgdata_cspc |
|
|
2210 | 2209 | dataOut.data_dc = avgdata_dc |
|
|
2211 | 2210 | dataOut.nIncohInt = self.incohInt |
|
|
2212 | 2211 | dataOut.data_outlier = self.nOutliers |
|
|
2213 | 2212 | dataOut.utctime = avgdatatime |
|
|
2214 | 2213 | dataOut.flagNoData = False |
|
|
2215 | 2214 | self.incohInt = 0 |
|
|
2216 | 2215 | self.nOutliers = 0 |
|
|
2217 | 2216 | self.__profIndex = 0 |
|
|
2218 | 2217 | dataOut.nProfilesByRange = self._nProfilesByRange |
|
|
2219 | 2218 | self._nProfilesByRange = numpy.zeros((1,len(dataOut.heightList))) |
|
|
2220 | 2219 | self._flagProfilesByRange = False |
|
|
2221 | ||
|
|
2220 | #print("IncohInt Done") | |
|
|
2222 | 2221 | return dataOut |
|
|
2223 | 2222 | |
|
|
2224 | 2223 | class dopplerFlip(Operation): |
|
|
2225 | 2224 | |
|
|
2226 | 2225 | def run(self, dataOut): |
|
|
2227 | 2226 | # arreglo 1: (num_chan, num_profiles, num_heights) |
|
|
2228 | 2227 | self.dataOut = dataOut |
|
|
2229 | 2228 | # JULIA-oblicua, indice 2 |
|
|
2230 | 2229 | # arreglo 2: (num_profiles, num_heights) |
|
|
2231 | 2230 | jspectra = self.dataOut.data_spc[2] |
|
|
2232 | 2231 | jspectra_tmp = numpy.zeros(jspectra.shape) |
|
|
2233 | 2232 | num_profiles = jspectra.shape[0] |
|
|
2234 | 2233 | freq_dc = int(num_profiles / 2) |
|
|
2235 | 2234 | # Flip con for |
|
|
2236 | 2235 | for j in range(num_profiles): |
|
|
2237 | 2236 | jspectra_tmp[num_profiles-j-1]= jspectra[j] |
|
|
2238 | 2237 | # Intercambio perfil de DC con perfil inmediato anterior |
|
|
2239 | 2238 | jspectra_tmp[freq_dc-1]= jspectra[freq_dc-1] |
|
|
2240 | 2239 | jspectra_tmp[freq_dc]= jspectra[freq_dc] |
|
|
2241 | 2240 | # canal modificado es re-escrito en el arreglo de canales |
|
|
2242 | 2241 | self.dataOut.data_spc[2] = jspectra_tmp |
|
|
2243 | 2242 | |
|
|
2244 | 2243 | return self.dataOut |
|
|
2245 | 2244 | |
|
|
2246 | 2245 | |
|
|
2247 | 2246 | |
|
|
2248 | 2247 | |
|
|
2249 | 2248 | |
|
|
2250 | 2249 | |
|
|
2251 | 2250 | class cleanJULIAInterf(Operation): |
|
|
2252 | 2251 | """ |
|
|
2253 | 2252 | OperaciΓ³n de prueba |
|
|
2254 | 2253 | """ |
|
|
2255 | 2254 | __slots__ =('heights_indx', 'repeats','span' ,'step', 'factor', 'idate', 'idxs','isConfig','minHrefN', 'maxHrefN') |
|
|
2256 | 2255 | def __init__(self): |
|
|
2257 | 2256 | self.repeats = 0 |
|
|
2258 | 2257 | self.factor=1 |
|
|
2259 | 2258 | self.isConfig = False |
|
|
2260 | 2259 | self.idxs = None |
|
|
2261 | 2260 | self.heights_indx = None |
|
|
2262 | 2261 | |
|
|
2263 | 2262 | def setup(self, dataOutHeightsList, heightsList, span=10, repeats=0, step=0, idate=None, startH=None, endH=None, minHref=None, maxHref=None): |
|
|
2264 | 2263 | totalHeihtList = dataOutHeightsList |
|
|
2265 | 2264 | heights = [float(hei) for hei in heightsList] |
|
|
2266 | 2265 | for r in range(repeats): |
|
|
2267 | 2266 | heights += [ (h+(step*(r+1))) for h in heights] |
|
|
2268 | 2267 | #print(heights) |
|
|
2269 | 2268 | self.heights_indx = [getHei_index(h,h,totalHeihtList)[0] for h in heights] |
|
|
2270 | 2269 | |
|
|
2271 | 2270 | self.minHrefN, self.maxHrefN = getHei_index(minHref,maxHref,totalHeihtList) |
|
|
2272 | 2271 | |
|
|
2273 | 2272 | |
|
|
2274 | 2273 | self.config = True |
|
|
2275 | 2274 | self.span = span |
|
|
2276 | 2275 | |
|
|
2277 | 2276 | def run(self, dataOut, heightsList, span=10, repeats=0, step=0, idate=None, startH=None, endH=None, minHref=None, maxHref=None): |
|
|
2278 | 2277 | |
|
|
2279 | 2278 | |
|
|
2280 | 2279 | self.dataOut = dataOut |
|
|
2281 | 2280 | startTime = datetime.datetime.combine(idate,startH) |
|
|
2282 | 2281 | endTime = datetime.datetime.combine(idate,endH) |
|
|
2283 | 2282 | currentTime = datetime.datetime.fromtimestamp(self.dataOut.utctime) |
|
|
2284 | 2283 | |
|
|
2285 | 2284 | if currentTime < startTime or currentTime > endTime: |
|
|
2286 | 2285 | return self.dataOut |
|
|
2287 | 2286 | |
|
|
2288 | 2287 | if not self.isConfig: |
|
|
2289 | 2288 | self.setup(self.dataOut.heightList,heightsList, span=span, repeats=repeats, step=step, idate=idate, startH=startH, endH=endH, minHref=minHref, maxHref=maxHref ) |
|
|
2290 | 2289 | |
|
|
2291 | 2290 | for ch in range(self.dataOut.data_spc.shape[0]): |
|
|
2292 | 2291 | i = 0 |
|
|
2293 | 2292 | N_ref = self.dataOut.data_spc[ch, :, self.minHrefN: self.maxHrefN].mean() |
|
|
2294 | 2293 | mn = self.heights_indx[-1] - self.span/2 |
|
|
2295 | 2294 | mx = self.heights_indx[-1] + self.span/2 |
|
|
2296 | 2295 | J_lev = self.dataOut.data_spc[ch, :, mn: mx].mean() - N_ref |
|
|
2297 | 2296 | |
|
|
2298 | 2297 | for hei in self.heights_indx: |
|
|
2299 | 2298 | h = hei - 1 |
|
|
2300 | 2299 | mn_i = hei - self.span/2 |
|
|
2301 | 2300 | mx_i = hei + self.span/2 |
|
|
2302 | 2301 | self.dataOut.data_spc[ch, :,mn_i:mx_i ] -= J_lev |
|
|
2303 | 2302 | i += 1 |
|
|
2304 | 2303 | |
|
|
2305 | 2304 | |
|
|
2306 | 2305 | return self.dataOut No newline at end of file |
| @@ -1,3259 +1,3249 | |||
|
|
1 | 1 | import sys |
|
|
2 | 2 | import numpy,math |
|
|
3 | 3 | from scipy import interpolate |
|
|
4 | 4 | from schainpy.model.proc.jroproc_base import ProcessingUnit, Operation, MPDecorator |
|
|
5 | 5 | from schainpy.model.data.jrodata import Voltage,hildebrand_sekhon |
|
|
6 | 6 | from schainpy.utils import log |
|
|
7 | 7 | from schainpy.model.io.utilsIO import getHei_index |
|
|
8 | 8 | from time import time |
|
|
9 | 9 | import datetime |
|
|
10 | 10 | import numpy |
|
|
11 | 11 | #import copy |
|
|
12 | 12 | from schainpy.model.data import _noise |
|
|
13 | 13 | |
|
|
14 | 14 | from matplotlib import pyplot as plt |
|
|
15 | 15 | |
|
|
16 | 16 | class VoltageProc(ProcessingUnit): |
|
|
17 | 17 | |
|
|
18 | 18 | def __init__(self): |
|
|
19 | 19 | |
|
|
20 | 20 | ProcessingUnit.__init__(self) |
|
|
21 | 21 | |
|
|
22 | 22 | self.dataOut = Voltage() |
|
|
23 | 23 | self.flip = 1 |
|
|
24 | 24 | self.setupReq = False |
|
|
25 | 25 | |
|
|
26 | 26 | def run(self): |
|
|
27 | 27 | #print("running volt proc") |
|
|
28 | 28 | |
|
|
29 | 29 | if self.dataIn.type == 'AMISR': |
|
|
30 | 30 | self.__updateObjFromAmisrInput() |
|
|
31 | 31 | |
|
|
32 | 32 | if self.dataOut.buffer_empty: |
|
|
33 | 33 | if self.dataIn.type == 'Voltage': |
|
|
34 | 34 | self.dataOut.copy(self.dataIn) |
|
|
35 | 35 | self.dataOut.radarControllerHeaderObj = self.dataIn.radarControllerHeaderObj.copy() |
|
|
36 | 36 | self.dataOut.ippSeconds = self.dataIn.ippSeconds |
|
|
37 | 37 | self.dataOut.ipp = self.dataIn.ipp |
|
|
38 | 38 | |
|
|
39 | 39 | #update Processing Header: |
|
|
40 | 40 | self.dataOut.processingHeaderObj.heightList = self.dataOut.heightList |
|
|
41 | 41 | self.dataOut.processingHeaderObj.ipp = self.dataOut.ipp |
|
|
42 | 42 | self.dataOut.processingHeaderObj.nCohInt = self.dataOut.nCohInt |
|
|
43 | 43 | self.dataOut.processingHeaderObj.dtype = self.dataOut.type |
|
|
44 | 44 | self.dataOut.processingHeaderObj.channelList = self.dataOut.channelList |
|
|
45 | 45 | self.dataOut.processingHeaderObj.azimuthList = self.dataOut.azimuthList |
|
|
46 | 46 | self.dataOut.processingHeaderObj.elevationList = self.dataOut.elevationList |
|
|
47 | 47 | self.dataOut.processingHeaderObj.codeList = self.dataOut.nChannels |
|
|
48 | 48 | self.dataOut.processingHeaderObj.heightList = self.dataOut.heightList |
|
|
49 | 49 | self.dataOut.processingHeaderObj.heightResolution = self.dataOut.heightList[1] - self.dataOut.heightList[0] |
|
|
50 | 50 | |
|
|
51 | 51 | |
|
|
52 | 52 | |
|
|
53 | 53 | def __updateObjFromAmisrInput(self): |
|
|
54 | 54 | |
|
|
55 | 55 | self.dataOut.timeZone = self.dataIn.timeZone |
|
|
56 | 56 | self.dataOut.dstFlag = self.dataIn.dstFlag |
|
|
57 | 57 | self.dataOut.errorCount = self.dataIn.errorCount |
|
|
58 | 58 | self.dataOut.useLocalTime = self.dataIn.useLocalTime |
|
|
59 | 59 | |
|
|
60 | 60 | self.dataOut.flagNoData = self.dataIn.flagNoData |
|
|
61 | 61 | self.dataOut.data = self.dataIn.data |
|
|
62 | 62 | self.dataOut.utctime = self.dataIn.utctime |
|
|
63 | 63 | self.dataOut.channelList = self.dataIn.channelList |
|
|
64 | 64 | #self.dataOut.timeInterval = self.dataIn.timeInterval |
|
|
65 | 65 | self.dataOut.heightList = self.dataIn.heightList |
|
|
66 | 66 | self.dataOut.nProfiles = self.dataIn.nProfiles |
|
|
67 | 67 | |
|
|
68 | 68 | self.dataOut.nCohInt = self.dataIn.nCohInt |
|
|
69 | 69 | self.dataOut.ippSeconds = self.dataIn.ippSeconds |
|
|
70 | 70 | self.dataOut.frequency = self.dataIn.frequency |
|
|
71 | 71 | |
|
|
72 | 72 | self.dataOut.azimuth = self.dataIn.azimuth |
|
|
73 | 73 | self.dataOut.zenith = self.dataIn.zenith |
|
|
74 | 74 | |
|
|
75 | 75 | self.dataOut.beam.codeList = self.dataIn.beam.codeList |
|
|
76 | 76 | self.dataOut.beam.azimuthList = self.dataIn.beam.azimuthList |
|
|
77 | 77 | self.dataOut.beam.zenithList = self.dataIn.beam.zenithList |
|
|
78 | 78 | |
|
|
79 | 79 | |
|
|
80 | 80 | class selectChannels(Operation): |
|
|
81 | 81 | |
|
|
82 | 82 | def run(self, dataOut, channelList=[]): |
|
|
83 | 83 | |
|
|
84 | 84 | if isinstance(channelList, int): |
|
|
85 | 85 | channelList = [channelList] |
|
|
86 | 86 | |
|
|
87 | 87 | self.channelList = channelList |
|
|
88 | 88 | if len(self.channelList) == 0: |
|
|
89 | 89 | print("Missing channelList") |
|
|
90 | 90 | return dataOut |
|
|
91 | 91 | channelIndexList = [] |
|
|
92 | 92 | if not dataOut.buffer_empty: # cuando se usa proc volts como buffer de datos |
|
|
93 | 93 | return dataOut |
|
|
94 | 94 | #print("channel List: ", dataOut.channelList) |
|
|
95 | 95 | if type(dataOut.channelList) is not list: #leer array desde HDF5 |
|
|
96 | 96 | try: |
|
|
97 | 97 | dataOut.channelList = dataOut.channelList.tolist() |
|
|
98 | 98 | except Exception as e: |
|
|
99 | 99 | print("Select Channels: ",e) |
|
|
100 | 100 | for channel in self.channelList: |
|
|
101 | 101 | if channel not in dataOut.channelList: |
|
|
102 | 102 | raise ValueError("Channel %d is not in %s" %(channel, str(dataOut.channelList))) |
|
|
103 | 103 | |
|
|
104 | 104 | index = dataOut.channelList.index(channel) |
|
|
105 | 105 | channelIndexList.append(index) |
|
|
106 | 106 | |
|
|
107 | 107 | dataOut = self.selectChannelsByIndex(dataOut,channelIndexList) |
|
|
108 | 108 | |
|
|
109 | 109 | #update Processing Header: |
|
|
110 | 110 | dataOut.processingHeaderObj.channelList = dataOut.channelList |
|
|
111 | 111 | dataOut.processingHeaderObj.elevationList = dataOut.elevationList |
|
|
112 | 112 | dataOut.processingHeaderObj.azimuthList = dataOut.azimuthList |
|
|
113 | 113 | dataOut.processingHeaderObj.codeList = dataOut.codeList |
|
|
114 | 114 | dataOut.processingHeaderObj.nChannels = len(dataOut.channelList) |
|
|
115 | 115 | |
|
|
116 | 116 | return dataOut |
|
|
117 | 117 | |
|
|
118 | 118 | def selectChannelsByIndex(self, dataOut, channelIndexList): |
|
|
119 | 119 | """ |
|
|
120 | 120 | Selecciona un bloque de datos en base a canales segun el channelIndexList |
|
|
121 | 121 | |
|
|
122 | 122 | Input: |
|
|
123 | 123 | channelIndexList : lista sencilla de canales a seleccionar por ej. [2,3,7] |
|
|
124 | 124 | |
|
|
125 | 125 | Affected: |
|
|
126 | 126 | dataOut.data |
|
|
127 | 127 | dataOut.channelIndexList |
|
|
128 | 128 | dataOut.nChannels |
|
|
129 | 129 | dataOut.m_ProcessingHeader.totalSpectra |
|
|
130 | 130 | dataOut.systemHeaderObj.numChannels |
|
|
131 | 131 | dataOut.m_ProcessingHeader.blockSize |
|
|
132 | 132 | |
|
|
133 | 133 | Return: |
|
|
134 | 134 | None |
|
|
135 | 135 | """ |
|
|
136 | 136 | #print("selectChannelsByIndex") |
|
|
137 | 137 | # for channelIndex in channelIndexList: |
|
|
138 | 138 | # if channelIndex not in dataOut.channelIndexList: |
|
|
139 | 139 | # raise ValueError("The value %d in channelIndexList is not valid" %channelIndex) |
|
|
140 | 140 | |
|
|
141 | 141 | if dataOut.type == 'Voltage': |
|
|
142 | 142 | if dataOut.flagDataAsBlock: |
|
|
143 | 143 | """ |
|
|
144 | 144 | Si la data es obtenida por bloques, dimension = [nChannels, nProfiles, nHeis] |
|
|
145 | 145 | """ |
|
|
146 | 146 | data = dataOut.data[channelIndexList,:,:] |
|
|
147 | 147 | else: |
|
|
148 | 148 | data = dataOut.data[channelIndexList,:] |
|
|
149 | 149 | |
|
|
150 | 150 | dataOut.data = data |
|
|
151 | 151 | # dataOut.channelList = [dataOut.channelList[i] for i in channelIndexList] |
|
|
152 | 152 | dataOut.channelList = [n for n in range(len(channelIndexList))] |
|
|
153 | 153 | |
|
|
154 | 154 | elif dataOut.type == 'Spectra': |
|
|
155 | 155 | if hasattr(dataOut, 'data_spc'): |
|
|
156 | 156 | if dataOut.data_spc is None: |
|
|
157 | 157 | raise ValueError("data_spc is None") |
|
|
158 | 158 | return dataOut |
|
|
159 | 159 | else: |
|
|
160 | 160 | data_spc = dataOut.data_spc[channelIndexList, :] |
|
|
161 | 161 | dataOut.data_spc = data_spc |
|
|
162 | 162 | |
|
|
163 | 163 | # if hasattr(dataOut, 'data_dc') :# and |
|
|
164 | 164 | # if dataOut.data_dc is None: |
|
|
165 | 165 | # raise ValueError("data_dc is None") |
|
|
166 | 166 | # return dataOut |
|
|
167 | 167 | # else: |
|
|
168 | 168 | # data_dc = dataOut.data_dc[channelIndexList, :] |
|
|
169 | 169 | # dataOut.data_dc = data_dc |
|
|
170 | 170 | # dataOut.channelList = [dataOut.channelList[i] for i in channelIndexList] |
|
|
171 | 171 | dataOut.channelList = channelIndexList |
|
|
172 | 172 | dataOut = self.__selectPairsByChannel(dataOut,channelIndexList) |
|
|
173 | 173 | |
|
|
174 | 174 | # channelIndexList = numpy.asarray(channelIndexList) |
|
|
175 | 175 | dataOut.elevationList = numpy.asarray(dataOut.elevationList) |
|
|
176 | 176 | dataOut.azimuthList = numpy.asarray(dataOut.azimuthList) |
|
|
177 | 177 | dataOut.codeList = numpy.asarray(dataOut.codeList) |
|
|
178 | 178 | if (len(dataOut.elevationList) > 0): |
|
|
179 | 179 | dataOut.elevationList = dataOut.elevationList[channelIndexList] |
|
|
180 | 180 | dataOut.azimuthList = dataOut.azimuthList[channelIndexList] |
|
|
181 | 181 | dataOut.codeList = dataOut.codeList[channelIndexList] |
|
|
182 | 182 | |
|
|
183 | 183 | return dataOut |
|
|
184 | 184 | |
|
|
185 | 185 | def __selectPairsByChannel(self, dataOut, channelList=None): |
|
|
186 | 186 | #print("__selectPairsByChannel") |
|
|
187 | 187 | if channelList == None: |
|
|
188 | 188 | return |
|
|
189 | 189 | |
|
|
190 | 190 | pairsIndexListSelected = [] |
|
|
191 | 191 | for pairIndex in dataOut.pairsIndexList: |
|
|
192 | 192 | # First pair |
|
|
193 | 193 | if dataOut.pairsList[pairIndex][0] not in channelList: |
|
|
194 | 194 | continue |
|
|
195 | 195 | # Second pair |
|
|
196 | 196 | if dataOut.pairsList[pairIndex][1] not in channelList: |
|
|
197 | 197 | continue |
|
|
198 | 198 | |
|
|
199 | 199 | pairsIndexListSelected.append(pairIndex) |
|
|
200 | 200 | if not pairsIndexListSelected: |
|
|
201 | 201 | dataOut.data_cspc = None |
|
|
202 | 202 | dataOut.pairsList = [] |
|
|
203 | 203 | return |
|
|
204 | 204 | |
|
|
205 | 205 | dataOut.data_cspc = dataOut.data_cspc[pairsIndexListSelected] |
|
|
206 | 206 | dataOut.pairsList = [dataOut.pairsList[i] |
|
|
207 | 207 | for i in pairsIndexListSelected] |
|
|
208 | 208 | |
|
|
209 | 209 | return dataOut |
|
|
210 | 210 | |
|
|
211 | 211 | class selectHeights(Operation): |
|
|
212 | 212 | |
|
|
213 | 213 | def run(self, dataOut, minHei=None, maxHei=None, minIndex=None, maxIndex=None): |
|
|
214 | 214 | """ |
|
|
215 | 215 | Selecciona un bloque de datos en base a un grupo de valores de alturas segun el rango |
|
|
216 | 216 | minHei <= height <= maxHei |
|
|
217 | 217 | |
|
|
218 | 218 | Input: |
|
|
219 | 219 | minHei : valor minimo de altura a considerar |
|
|
220 | 220 | maxHei : valor maximo de altura a considerar |
|
|
221 | 221 | |
|
|
222 | 222 | Affected: |
|
|
223 | 223 | Indirectamente son cambiados varios valores a travez del metodo selectHeightsByIndex |
|
|
224 | 224 | |
|
|
225 | 225 | Return: |
|
|
226 | 226 | 1 si el metodo se ejecuto con exito caso contrario devuelve 0 |
|
|
227 | 227 | """ |
|
|
228 | 228 | |
|
|
229 | 229 | self.dataOut = dataOut |
|
|
230 | 230 | |
|
|
231 | 231 | if minHei and maxHei: |
|
|
232 | 232 | |
|
|
233 | 233 | if (minHei < dataOut.heightList[0]): |
|
|
234 | 234 | minHei = dataOut.heightList[0] |
|
|
235 | 235 | |
|
|
236 | 236 | if (maxHei > dataOut.heightList[-1]): |
|
|
237 | 237 | maxHei = dataOut.heightList[-1] |
|
|
238 | 238 | |
|
|
239 | 239 | minIndex = 0 |
|
|
240 | 240 | maxIndex = 0 |
|
|
241 | 241 | heights = dataOut.heightList |
|
|
242 | 242 | |
|
|
243 | 243 | inda = numpy.where(heights >= minHei) |
|
|
244 | 244 | indb = numpy.where(heights <= maxHei) |
|
|
245 | 245 | |
|
|
246 | 246 | try: |
|
|
247 | 247 | minIndex = inda[0][0] |
|
|
248 | 248 | except: |
|
|
249 | 249 | minIndex = 0 |
|
|
250 | 250 | |
|
|
251 | 251 | try: |
|
|
252 | 252 | maxIndex = indb[0][-1] |
|
|
253 | 253 | except: |
|
|
254 | 254 | maxIndex = len(heights) |
|
|
255 | 255 | |
|
|
256 | 256 | self.selectHeightsByIndex(minIndex, maxIndex) |
|
|
257 | 257 | |
|
|
258 | 258 | #update Processing Header: |
|
|
259 | 259 | dataOut.processingHeaderObj.heightList = dataOut.heightList |
|
|
260 | 260 | |
|
|
261 | 261 | |
|
|
262 | 262 | |
|
|
263 | 263 | return dataOut |
|
|
264 | 264 | |
|
|
265 | 265 | def selectHeightsByIndex(self, minIndex, maxIndex): |
|
|
266 | 266 | """ |
|
|
267 | 267 | Selecciona un bloque de datos en base a un grupo indices de alturas segun el rango |
|
|
268 | 268 | minIndex <= index <= maxIndex |
|
|
269 | 269 | |
|
|
270 | 270 | Input: |
|
|
271 | 271 | minIndex : valor de indice minimo de altura a considerar |
|
|
272 | 272 | maxIndex : valor de indice maximo de altura a considerar |
|
|
273 | 273 | |
|
|
274 | 274 | Affected: |
|
|
275 | 275 | self.dataOut.data |
|
|
276 | 276 | self.dataOut.heightList |
|
|
277 | 277 | |
|
|
278 | 278 | Return: |
|
|
279 | 279 | 1 si el metodo se ejecuto con exito caso contrario devuelve 0 |
|
|
280 | 280 | """ |
|
|
281 | 281 | |
|
|
282 | 282 | if self.dataOut.type == 'Voltage': |
|
|
283 | 283 | if (minIndex < 0) or (minIndex > maxIndex): |
|
|
284 | 284 | raise ValueError("Height index range (%d,%d) is not valid" % (minIndex, maxIndex)) |
|
|
285 | 285 | |
|
|
286 | 286 | if (maxIndex >= self.dataOut.nHeights): |
|
|
287 | 287 | maxIndex = self.dataOut.nHeights |
|
|
288 | 288 | |
|
|
289 | 289 | #voltage |
|
|
290 | 290 | if self.dataOut.flagDataAsBlock: |
|
|
291 | 291 | """ |
|
|
292 | 292 | Si la data es obtenida por bloques, dimension = [nChannels, nProfiles, nHeis] |
|
|
293 | 293 | """ |
|
|
294 | 294 | data = self.dataOut.data[:,:, minIndex:maxIndex] |
|
|
295 | 295 | else: |
|
|
296 | 296 | data = self.dataOut.data[:, minIndex:maxIndex] |
|
|
297 | 297 | |
|
|
298 | 298 | # firstHeight = self.dataOut.heightList[minIndex] |
|
|
299 | 299 | |
|
|
300 | 300 | self.dataOut.data = data |
|
|
301 | 301 | self.dataOut.heightList = self.dataOut.heightList[minIndex:maxIndex] |
|
|
302 | 302 | |
|
|
303 | 303 | if self.dataOut.nHeights <= 1: |
|
|
304 | 304 | raise ValueError("selectHeights: Too few heights. Current number of heights is %d" %(self.dataOut.nHeights)) |
|
|
305 | 305 | elif self.dataOut.type == 'Spectra': |
|
|
306 | 306 | if (minIndex < 0) or (minIndex > maxIndex): |
|
|
307 | 307 | raise ValueError("Error selecting heights: Index range (%d,%d) is not valid" % ( |
|
|
308 | 308 | minIndex, maxIndex)) |
|
|
309 | 309 | |
|
|
310 | 310 | if (maxIndex >= self.dataOut.nHeights): |
|
|
311 | 311 | maxIndex = self.dataOut.nHeights - 1 |
|
|
312 | 312 | |
|
|
313 | 313 | # Spectra |
|
|
314 | 314 | data_spc = self.dataOut.data_spc[:, :, minIndex:maxIndex + 1] |
|
|
315 | 315 | |
|
|
316 | 316 | data_cspc = None |
|
|
317 | 317 | if self.dataOut.data_cspc is not None: |
|
|
318 | 318 | data_cspc = self.dataOut.data_cspc[:, :, minIndex:maxIndex + 1] |
|
|
319 | 319 | |
|
|
320 | 320 | data_dc = None |
|
|
321 | 321 | if self.dataOut.data_dc is not None: |
|
|
322 | 322 | data_dc = self.dataOut.data_dc[:, minIndex:maxIndex + 1] |
|
|
323 | 323 | |
|
|
324 | 324 | self.dataOut.data_spc = data_spc |
|
|
325 | 325 | self.dataOut.data_cspc = data_cspc |
|
|
326 | 326 | self.dataOut.data_dc = data_dc |
|
|
327 | 327 | |
|
|
328 | 328 | self.dataOut.heightList = self.dataOut.heightList[minIndex:maxIndex + 1] |
|
|
329 | 329 | |
|
|
330 | 330 | return 1 |
|
|
331 | 331 | |
|
|
332 | 332 | |
|
|
333 | 333 | class filterByHeights(Operation): |
|
|
334 | 334 | ifConfig=False |
|
|
335 | 335 | deltaHeight = None |
|
|
336 | 336 | newdelta=None |
|
|
337 | 337 | newheights=None |
|
|
338 | 338 | r=None |
|
|
339 | 339 | h0=None |
|
|
340 | 340 | nHeights=None |
|
|
341 | 341 | def run(self, dataOut, window): |
|
|
342 | 342 | |
|
|
343 | 343 | |
|
|
344 | 344 | # print("1",dataOut.data.shape) |
|
|
345 | 345 | # print(dataOut.nHeights) |
|
|
346 | 346 | if window == None: |
|
|
347 | 347 | window = (dataOut.radarControllerHeaderObj.txA/dataOut.radarControllerHeaderObj.nBaud) / self.deltaHeight |
|
|
348 | 348 | |
|
|
349 | 349 | if not self.ifConfig: #and dataOut.useInputBuffer: |
|
|
350 | 350 | self.deltaHeight = dataOut.heightList[1] - dataOut.heightList[0] |
|
|
351 | 351 | self.ifConfig = True |
|
|
352 | 352 | self.newdelta = self.deltaHeight * window |
|
|
353 | 353 | self.r = dataOut.nHeights % window |
|
|
354 | 354 | self.newheights = (dataOut.nHeights-self.r)/window |
|
|
355 | 355 | self.h0 = dataOut.heightList[0] |
|
|
356 | 356 | self.nHeights = dataOut.nHeights |
|
|
357 | 357 | if self.newheights <= 1: |
|
|
358 | 358 | raise ValueError("filterByHeights: Too few heights. Current number of heights is %d and window is %d" %(dataOut.nHeights, window)) |
|
|
359 | 359 | |
|
|
360 | 360 | if dataOut.flagDataAsBlock: |
|
|
361 | 361 | """ |
|
|
362 | 362 | Si la data es obtenida por bloques, dimension = [nChannels, nProfiles, nHeis] |
|
|
363 | 363 | """ |
|
|
364 | 364 | buffer = dataOut.data[:, :, 0:int(self.nHeights-self.r)] |
|
|
365 | 365 | buffer = buffer.reshape(dataOut.nChannels, dataOut.nProfiles, int(self.nHeights/window), window) |
|
|
366 | 366 | buffer = numpy.sum(buffer,3) |
|
|
367 | 367 | |
|
|
368 | 368 | else: |
|
|
369 | 369 | buffer = dataOut.data[:,0:int(self.nHeights-self.r)] |
|
|
370 | 370 | buffer = buffer.reshape(dataOut.nChannels,int(self.nHeights/window),int(window)) |
|
|
371 | 371 | buffer = numpy.sum(buffer,2) |
|
|
372 | 372 | |
|
|
373 | 373 | dataOut.data = buffer |
|
|
374 | 374 | dataOut.heightList = self.h0 + numpy.arange( self.newheights )*self.newdelta |
|
|
375 | 375 | dataOut.windowOfFilter = window |
|
|
376 | 376 | |
|
|
377 | 377 | #update Processing Header: |
|
|
378 | 378 | dataOut.processingHeaderObj.heightList = dataOut.heightList |
|
|
379 | 379 | dataOut.processingHeaderObj.nWindows = window |
|
|
380 | 380 | |
|
|
381 | 381 | return dataOut |
|
|
382 | 382 | |
|
|
383 | 383 | |
|
|
384 | 384 | |
|
|
385 | 385 | class setH0(Operation): |
|
|
386 | 386 | |
|
|
387 | 387 | def run(self, dataOut, h0, deltaHeight = None): |
|
|
388 | 388 | |
|
|
389 | 389 | if not deltaHeight: |
|
|
390 | 390 | deltaHeight = dataOut.heightList[1] - dataOut.heightList[0] |
|
|
391 | 391 | |
|
|
392 | 392 | nHeights = dataOut.nHeights |
|
|
393 | 393 | |
|
|
394 | 394 | newHeiRange = h0 + numpy.arange(nHeights)*deltaHeight |
|
|
395 | 395 | |
|
|
396 | 396 | dataOut.heightList = newHeiRange |
|
|
397 | 397 | |
|
|
398 | 398 | #update Processing Header: |
|
|
399 | 399 | dataOut.processingHeaderObj.heightList = dataOut.heightList |
|
|
400 | 400 | |
|
|
401 | 401 | return dataOut |
|
|
402 | 402 | |
|
|
403 | 403 | |
|
|
404 | 404 | class deFlip(Operation): |
|
|
405 | 405 | |
|
|
406 | 406 | def run(self, dataOut, channelList = []): |
|
|
407 | 407 | |
|
|
408 | 408 | data = dataOut.data.copy() |
|
|
409 | 409 | |
|
|
410 | 410 | if dataOut.flagDataAsBlock: |
|
|
411 | 411 | flip = self.flip |
|
|
412 | 412 | profileList = list(range(dataOut.nProfiles)) |
|
|
413 | 413 | |
|
|
414 | 414 | if not channelList: |
|
|
415 | 415 | for thisProfile in profileList: |
|
|
416 | 416 | data[:,thisProfile,:] = data[:,thisProfile,:]*flip |
|
|
417 | 417 | flip *= -1.0 |
|
|
418 | 418 | else: |
|
|
419 | 419 | for thisChannel in channelList: |
|
|
420 | 420 | if thisChannel not in dataOut.channelList: |
|
|
421 | 421 | continue |
|
|
422 | 422 | |
|
|
423 | 423 | for thisProfile in profileList: |
|
|
424 | 424 | data[thisChannel,thisProfile,:] = data[thisChannel,thisProfile,:]*flip |
|
|
425 | 425 | flip *= -1.0 |
|
|
426 | 426 | |
|
|
427 | 427 | self.flip = flip |
|
|
428 | 428 | |
|
|
429 | 429 | else: |
|
|
430 | 430 | if not channelList: |
|
|
431 | 431 | data[:,:] = data[:,:]*self.flip |
|
|
432 | 432 | else: |
|
|
433 | 433 | for thisChannel in channelList: |
|
|
434 | 434 | if thisChannel not in dataOut.channelList: |
|
|
435 | 435 | continue |
|
|
436 | 436 | |
|
|
437 | 437 | data[thisChannel,:] = data[thisChannel,:]*self.flip |
|
|
438 | 438 | |
|
|
439 | 439 | self.flip *= -1. |
|
|
440 | 440 | |
|
|
441 | 441 | dataOut.data = data |
|
|
442 | 442 | |
|
|
443 | 443 | return dataOut |
|
|
444 | 444 | |
|
|
445 | 445 | |
|
|
446 | 446 | class setAttribute(Operation): |
|
|
447 | 447 | ''' |
|
|
448 | 448 | Set an arbitrary attribute(s) to dataOut |
|
|
449 | 449 | ''' |
|
|
450 | 450 | |
|
|
451 | 451 | def __init__(self): |
|
|
452 | 452 | |
|
|
453 | 453 | Operation.__init__(self) |
|
|
454 | 454 | self._ready = False |
|
|
455 | 455 | |
|
|
456 | 456 | def run(self, dataOut, **kwargs): |
|
|
457 | 457 | |
|
|
458 | 458 | for key, value in kwargs.items(): |
|
|
459 | 459 | setattr(dataOut, key, value) |
|
|
460 | 460 | |
|
|
461 | 461 | return dataOut |
|
|
462 | 462 | |
|
|
463 | 463 | |
|
|
464 | 464 | @MPDecorator |
|
|
465 | 465 | class printAttribute(Operation): |
|
|
466 | 466 | ''' |
|
|
467 | 467 | Print an arbitrary attribute of dataOut |
|
|
468 | 468 | ''' |
|
|
469 | 469 | |
|
|
470 | 470 | def __init__(self): |
|
|
471 | 471 | |
|
|
472 | 472 | Operation.__init__(self) |
|
|
473 | 473 | |
|
|
474 | 474 | def run(self, dataOut, attributes): |
|
|
475 | 475 | |
|
|
476 | 476 | if isinstance(attributes, str): |
|
|
477 | 477 | attributes = [attributes] |
|
|
478 | 478 | for attr in attributes: |
|
|
479 | 479 | if hasattr(dataOut, attr): |
|
|
480 | 480 | log.log(getattr(dataOut, attr), attr) |
|
|
481 | 481 | |
|
|
482 | 482 | class cleanHeightsInterf(Operation): |
|
|
483 | 483 | __slots__ =('heights_indx', 'repeats', 'step', 'factor', 'idate', 'idxs','config','wMask') |
|
|
484 | 484 | def __init__(self): |
|
|
485 | 485 | self.repeats = 0 |
|
|
486 | 486 | self.factor=1 |
|
|
487 | 487 | self.wMask = None |
|
|
488 | 488 | self.config = False |
|
|
489 | 489 | self.idxs = None |
|
|
490 | 490 | self.heights_indx = None |
|
|
491 | 491 | |
|
|
492 | 492 | def run(self, dataOut, heightsList, repeats=0, step=0, factor=1, idate=None, startH=None, endH=None): |
|
|
493 | 493 | |
|
|
494 | 494 | #print(dataOut.data.shape) |
|
|
495 | 495 | |
|
|
496 | 496 | startTime = datetime.datetime.combine(idate,startH) |
|
|
497 | 497 | endTime = datetime.datetime.combine(idate,endH) |
|
|
498 | 498 | currentTime = datetime.datetime.fromtimestamp(dataOut.utctime) |
|
|
499 | 499 | |
|
|
500 | 500 | if currentTime < startTime or currentTime > endTime: |
|
|
501 | 501 | return dataOut |
|
|
502 | 502 | if not self.config: |
|
|
503 | 503 | |
|
|
504 | 504 | #print(wMask) |
|
|
505 | 505 | heights = [float(hei) for hei in heightsList] |
|
|
506 | 506 | for r in range(repeats): |
|
|
507 | 507 | heights += [ (h+(step*(r+1))) for h in heights] |
|
|
508 | 508 | #print(heights) |
|
|
509 | 509 | heiList = dataOut.heightList |
|
|
510 | 510 | self.heights_indx = [getHei_index(h,h,heiList)[0] for h in heights] |
|
|
511 | 511 | |
|
|
512 | 512 | self.wMask = numpy.asarray(factor) |
|
|
513 | 513 | self.wMask = numpy.tile(self.wMask,(repeats+2)) |
|
|
514 | 514 | self.config = True |
|
|
515 | 515 | |
|
|
516 | 516 | """ |
|
|
517 | 517 | getNoisebyHildebrand(self, channel=None, ymin_index=None, ymax_index=None) |
|
|
518 | 518 | """ |
|
|
519 | 519 | #print(self.noise =10*numpy.log10(dataOut.getNoisebyHildebrand(ymin_index=self.min_ref, ymax_index=self.max_ref))) |
|
|
520 | 520 | |
|
|
521 | 521 | |
|
|
522 | 522 | for ch in range(dataOut.data.shape[0]): |
|
|
523 | 523 | i = 0 |
|
|
524 | 524 | |
|
|
525 | 525 | |
|
|
526 | 526 | for hei in self.heights_indx: |
|
|
527 | 527 | h = hei - 1 |
|
|
528 | 528 | |
|
|
529 | 529 | |
|
|
530 | 530 | if dataOut.data.ndim < 3: |
|
|
531 | 531 | module = numpy.absolute(dataOut.data[ch,h]) |
|
|
532 | 532 | prev_h1 = numpy.absolute(dataOut.data[ch,h-1]) |
|
|
533 | 533 | dataOut.data[ch,h] = (dataOut.data[ch,h])/module * prev_h1 |
|
|
534 | 534 | |
|
|
535 | 535 | #dataOut.data[ch,hei-1] = (dataOut.data[ch,hei-1])*self.wMask[i] |
|
|
536 | 536 | else: |
|
|
537 | 537 | module = numpy.absolute(dataOut.data[ch,:,h]) |
|
|
538 | 538 | prev_h1 = numpy.absolute(dataOut.data[ch,:,h-1]) |
|
|
539 | 539 | dataOut.data[ch,:,h] = (dataOut.data[ch,:,h])/module * prev_h1 |
|
|
540 | 540 | #dataOut.data[ch,:,hei-1] = (dataOut.data[ch,:,hei-1])*self.wMask[i] |
|
|
541 | 541 | #print("done") |
|
|
542 | 542 | i += 1 |
|
|
543 | 543 | |
|
|
544 | 544 | |
|
|
545 | 545 | return dataOut |
|
|
546 | 546 | |
|
|
547 | 547 | |
|
|
548 | 548 | |
|
|
549 | 549 | class interpolateHeights(Operation): |
|
|
550 | 550 | |
|
|
551 | 551 | def run(self, dataOut, topLim, botLim): |
|
|
552 | 552 | #69 al 72 para julia |
|
|
553 | 553 | #82-84 para meteoros |
|
|
554 | 554 | if len(numpy.shape(dataOut.data))==2: |
|
|
555 | 555 | sampInterp = (dataOut.data[:,botLim-1] + dataOut.data[:,topLim+1])/2 |
|
|
556 | 556 | sampInterp = numpy.transpose(numpy.tile(sampInterp,(topLim-botLim + 1,1))) |
|
|
557 | 557 | #dataOut.data[:,botLim:limSup+1] = sampInterp |
|
|
558 | 558 | dataOut.data[:,botLim:topLim+1] = sampInterp |
|
|
559 | 559 | else: |
|
|
560 | 560 | nHeights = dataOut.data.shape[2] |
|
|
561 | 561 | x = numpy.hstack((numpy.arange(botLim),numpy.arange(topLim+1,nHeights))) |
|
|
562 | 562 | y = dataOut.data[:,:,list(range(botLim))+list(range(topLim+1,nHeights))] |
|
|
563 | 563 | f = interpolate.interp1d(x, y, axis = 2) |
|
|
564 | 564 | xnew = numpy.arange(botLim,topLim+1) |
|
|
565 | 565 | ynew = f(xnew) |
|
|
566 | 566 | dataOut.data[:,:,botLim:topLim+1] = ynew |
|
|
567 | 567 | |
|
|
568 | 568 | return dataOut |
|
|
569 | 569 | |
|
|
570 | 570 | |
|
|
571 | 571 | class CohInt(Operation): |
|
|
572 | 572 | |
|
|
573 | 573 | isConfig = False |
|
|
574 | 574 | __profIndex = 0 |
|
|
575 | 575 | __byTime = False |
|
|
576 | 576 | __initime = None |
|
|
577 | 577 | __lastdatatime = None |
|
|
578 | 578 | __integrationtime = None |
|
|
579 | 579 | __buffer = None |
|
|
580 | 580 | __bufferStride = [] |
|
|
581 | 581 | __dataReady = False |
|
|
582 | 582 | __profIndexStride = 0 |
|
|
583 | 583 | __dataToPutStride = False |
|
|
584 | 584 | n = None |
|
|
585 | 585 | |
|
|
586 | 586 | def __init__(self, **kwargs): |
|
|
587 | 587 | |
|
|
588 | 588 | Operation.__init__(self, **kwargs) |
|
|
589 | 589 | |
|
|
590 | 590 | def setup(self, n=None, timeInterval=None, stride=None, overlapping=False, byblock=False): |
|
|
591 | 591 | """ |
|
|
592 | 592 | Set the parameters of the integration class. |
|
|
593 | 593 | |
|
|
594 | 594 | Inputs: |
|
|
595 | 595 | |
|
|
596 | 596 | n : Number of coherent integrations |
|
|
597 | 597 | timeInterval : Time of integration. If the parameter "n" is selected this one does not work |
|
|
598 | 598 | overlapping : |
|
|
599 | 599 | """ |
|
|
600 | 600 | |
|
|
601 | 601 | self.__initime = None |
|
|
602 | 602 | self.__lastdatatime = 0 |
|
|
603 | 603 | self.__buffer = None |
|
|
604 | 604 | self.__dataReady = False |
|
|
605 | 605 | self.byblock = byblock |
|
|
606 | 606 | self.stride = stride |
|
|
607 | 607 | |
|
|
608 | 608 | if n == None and timeInterval == None: |
|
|
609 | 609 | raise ValueError("n or timeInterval should be specified ...") |
|
|
610 | 610 | |
|
|
611 | 611 | if n != None: |
|
|
612 | 612 | self.n = n |
|
|
613 | 613 | self.__byTime = False |
|
|
614 | 614 | else: |
|
|
615 | 615 | self.__integrationtime = timeInterval #* 60. #if (type(timeInterval)!=integer) -> change this line |
|
|
616 | 616 | self.n = 9999 |
|
|
617 | 617 | self.__byTime = True |
|
|
618 | 618 | |
|
|
619 | 619 | if overlapping: |
|
|
620 | 620 | self.__withOverlapping = True |
|
|
621 | 621 | self.__buffer = None |
|
|
622 | 622 | else: |
|
|
623 | 623 | self.__withOverlapping = False |
|
|
624 | 624 | self.__buffer = 0 |
|
|
625 | 625 | |
|
|
626 | 626 | self.__profIndex = 0 |
|
|
627 | 627 | |
|
|
628 | 628 | def putData(self, data): |
|
|
629 | 629 | |
|
|
630 | 630 | """ |
|
|
631 | 631 | Add a profile to the __buffer and increase in one the __profileIndex |
|
|
632 | 632 | |
|
|
633 | 633 | """ |
|
|
634 | 634 | |
|
|
635 | 635 | if not self.__withOverlapping: |
|
|
636 | 636 | self.__buffer += data.copy() |
|
|
637 | 637 | self.__profIndex += 1 |
|
|
638 | 638 | return |
|
|
639 | 639 | |
|
|
640 | 640 | #Overlapping data |
|
|
641 | 641 | nChannels, nHeis = data.shape |
|
|
642 | 642 | data = numpy.reshape(data, (1, nChannels, nHeis)) |
|
|
643 | 643 | |
|
|
644 | 644 | #If the buffer is empty then it takes the data value |
|
|
645 | 645 | if self.__buffer is None: |
|
|
646 | 646 | self.__buffer = data |
|
|
647 | 647 | self.__profIndex += 1 |
|
|
648 | 648 | return |
|
|
649 | 649 | |
|
|
650 | 650 | #If the buffer length is lower than n then stakcing the data value |
|
|
651 | 651 | if self.__profIndex < self.n: |
|
|
652 | 652 | self.__buffer = numpy.vstack((self.__buffer, data)) |
|
|
653 | 653 | self.__profIndex += 1 |
|
|
654 | 654 | return |
|
|
655 | 655 | |
|
|
656 | 656 | #If the buffer length is equal to n then replacing the last buffer value with the data value |
|
|
657 | 657 | self.__buffer = numpy.roll(self.__buffer, -1, axis=0) |
|
|
658 | 658 | self.__buffer[self.n-1] = data |
|
|
659 | 659 | self.__profIndex = self.n |
|
|
660 | 660 | return |
|
|
661 | 661 | |
|
|
662 | 662 | |
|
|
663 | 663 | def pushData(self): |
|
|
664 | 664 | """ |
|
|
665 | 665 | Return the sum of the last profiles and the profiles used in the sum. |
|
|
666 | 666 | |
|
|
667 | 667 | Affected: |
|
|
668 | 668 | |
|
|
669 | 669 | self.__profileIndex |
|
|
670 | 670 | |
|
|
671 | 671 | """ |
|
|
672 | 672 | |
|
|
673 | 673 | if not self.__withOverlapping: |
|
|
674 | 674 | data = self.__buffer |
|
|
675 | 675 | n = self.__profIndex |
|
|
676 | 676 | |
|
|
677 | 677 | self.__buffer = 0 |
|
|
678 | 678 | self.__profIndex = 0 |
|
|
679 | 679 | |
|
|
680 | 680 | return data, n |
|
|
681 | 681 | |
|
|
682 | 682 | #Integration with Overlapping |
|
|
683 | 683 | data = numpy.sum(self.__buffer, axis=0) |
|
|
684 | 684 | # print data |
|
|
685 | 685 | # raise |
|
|
686 | 686 | n = self.__profIndex |
|
|
687 | 687 | |
|
|
688 | 688 | return data, n |
|
|
689 | 689 | |
|
|
690 | 690 | def byProfiles(self, data): |
|
|
691 | 691 | |
|
|
692 | 692 | self.__dataReady = False |
|
|
693 | 693 | avgdata = None |
|
|
694 | 694 | # n = None |
|
|
695 | 695 | # print data |
|
|
696 | 696 | # raise |
|
|
697 | 697 | self.putData(data) |
|
|
698 | 698 | |
|
|
699 | 699 | if self.__profIndex == self.n: |
|
|
700 | 700 | avgdata, n = self.pushData() |
|
|
701 | 701 | self.__dataReady = True |
|
|
702 | 702 | |
|
|
703 | 703 | return avgdata |
|
|
704 | 704 | |
|
|
705 | 705 | def byTime(self, data, datatime): |
|
|
706 | 706 | |
|
|
707 | 707 | self.__dataReady = False |
|
|
708 | 708 | avgdata = None |
|
|
709 | 709 | n = None |
|
|
710 | 710 | |
|
|
711 | 711 | self.putData(data) |
|
|
712 | 712 | |
|
|
713 | 713 | if (datatime - self.__initime) >= self.__integrationtime: |
|
|
714 | 714 | avgdata, n = self.pushData() |
|
|
715 | 715 | self.n = n |
|
|
716 | 716 | self.__dataReady = True |
|
|
717 | 717 | |
|
|
718 | 718 | return avgdata |
|
|
719 | 719 | |
|
|
720 | 720 | def integrateByStride(self, data, datatime): |
|
|
721 | 721 | # print data |
|
|
722 | 722 | if self.__profIndex == 0: |
|
|
723 | 723 | self.__buffer = [[data.copy(), datatime]] |
|
|
724 | 724 | else: |
|
|
725 | 725 | self.__buffer.append([data.copy(),datatime]) |
|
|
726 | 726 | self.__profIndex += 1 |
|
|
727 | 727 | self.__dataReady = False |
|
|
728 | 728 | |
|
|
729 | 729 | if self.__profIndex == self.n * self.stride : |
|
|
730 | 730 | self.__dataToPutStride = True |
|
|
731 | 731 | self.__profIndexStride = 0 |
|
|
732 | 732 | self.__profIndex = 0 |
|
|
733 | 733 | self.__bufferStride = [] |
|
|
734 | 734 | for i in range(self.stride): |
|
|
735 | 735 | current = self.__buffer[i::self.stride] |
|
|
736 | 736 | data = numpy.sum([t[0] for t in current], axis=0) |
|
|
737 | 737 | avgdatatime = numpy.average([t[1] for t in current]) |
|
|
738 | 738 | # print data |
|
|
739 | 739 | self.__bufferStride.append((data, avgdatatime)) |
|
|
740 | 740 | |
|
|
741 | 741 | if self.__dataToPutStride: |
|
|
742 | 742 | self.__dataReady = True |
|
|
743 | 743 | self.__profIndexStride += 1 |
|
|
744 | 744 | if self.__profIndexStride == self.stride: |
|
|
745 | 745 | self.__dataToPutStride = False |
|
|
746 | 746 | # print self.__bufferStride[self.__profIndexStride - 1] |
|
|
747 | 747 | # raise |
|
|
748 | 748 | return self.__bufferStride[self.__profIndexStride - 1] |
|
|
749 | 749 | |
|
|
750 | 750 | |
|
|
751 | 751 | return None, None |
|
|
752 | 752 | |
|
|
753 | 753 | def integrate(self, data, datatime=None): |
|
|
754 | 754 | |
|
|
755 | 755 | if self.__initime == None: |
|
|
756 | 756 | self.__initime = datatime |
|
|
757 | 757 | |
|
|
758 | 758 | if self.__byTime: |
|
|
759 | 759 | avgdata = self.byTime(data, datatime) |
|
|
760 | 760 | else: |
|
|
761 | 761 | avgdata = self.byProfiles(data) |
|
|
762 | 762 | |
|
|
763 | 763 | |
|
|
764 | 764 | self.__lastdatatime = datatime |
|
|
765 | 765 | |
|
|
766 | 766 | if avgdata is None: |
|
|
767 | 767 | return None, None |
|
|
768 | 768 | |
|
|
769 | 769 | avgdatatime = self.__initime |
|
|
770 | 770 | |
|
|
771 | 771 | deltatime = datatime - self.__lastdatatime |
|
|
772 | 772 | |
|
|
773 | 773 | if not self.__withOverlapping: |
|
|
774 | 774 | self.__initime = datatime |
|
|
775 | 775 | else: |
|
|
776 | 776 | self.__initime += deltatime |
|
|
777 | 777 | |
|
|
778 | 778 | return avgdata, avgdatatime |
|
|
779 | 779 | |
|
|
780 | 780 | def integrateByBlock(self, dataOut): |
|
|
781 | 781 | |
|
|
782 | 782 | times = int(dataOut.data.shape[1]/self.n) |
|
|
783 | 783 | avgdata = numpy.zeros((dataOut.nChannels, times, dataOut.nHeights), dtype=numpy.complex) |
|
|
784 | 784 | |
|
|
785 | 785 | id_min = 0 |
|
|
786 | 786 | id_max = self.n |
|
|
787 | 787 | |
|
|
788 | 788 | for i in range(times): |
|
|
789 | 789 | junk = dataOut.data[:,id_min:id_max,:] |
|
|
790 | 790 | avgdata[:,i,:] = junk.sum(axis=1) |
|
|
791 | 791 | id_min += self.n |
|
|
792 | 792 | id_max += self.n |
|
|
793 | 793 | |
|
|
794 | 794 | timeInterval = dataOut.ippSeconds*self.n |
|
|
795 | 795 | avgdatatime = (times - 1) * timeInterval + dataOut.utctime |
|
|
796 | 796 | self.__dataReady = True |
|
|
797 | 797 | return avgdata, avgdatatime |
|
|
798 | 798 | |
|
|
799 | 799 | def run(self, dataOut, n=None, timeInterval=None, stride=None, overlapping=False, byblock=False, **kwargs): |
|
|
800 | 800 | |
|
|
801 | 801 | if not self.isConfig: |
|
|
802 | 802 | self.setup(n=n, stride=stride, timeInterval=timeInterval, overlapping=overlapping, byblock=byblock, **kwargs) |
|
|
803 | 803 | self.isConfig = True |
|
|
804 | 804 | |
|
|
805 | 805 | if dataOut.flagDataAsBlock: |
|
|
806 | 806 | """ |
|
|
807 | 807 | Si la data es leida por bloques, dimension = [nChannels, nProfiles, nHeis] |
|
|
808 | 808 | """ |
|
|
809 | 809 | avgdata, avgdatatime = self.integrateByBlock(dataOut) |
|
|
810 | 810 | dataOut.nProfiles /= self.n |
|
|
811 | 811 | else: |
|
|
812 | 812 | if stride is None: |
|
|
813 | 813 | avgdata, avgdatatime = self.integrate(dataOut.data, dataOut.utctime) |
|
|
814 | 814 | else: |
|
|
815 | 815 | avgdata, avgdatatime = self.integrateByStride(dataOut.data, dataOut.utctime) |
|
|
816 | 816 | |
|
|
817 | 817 | |
|
|
818 | 818 | # dataOut.timeInterval *= n |
|
|
819 | 819 | dataOut.flagNoData = True |
|
|
820 | 820 | |
|
|
821 | 821 | if self.__dataReady: |
|
|
822 | 822 | dataOut.data = avgdata |
|
|
823 | 823 | if not dataOut.flagCohInt: |
|
|
824 | 824 | dataOut.nCohInt *= self.n |
|
|
825 | 825 | dataOut.flagCohInt = True |
|
|
826 | 826 | dataOut.utctime = avgdatatime |
|
|
827 | 827 | # print avgdata, avgdatatime |
|
|
828 | 828 | # raise |
|
|
829 | 829 | # dataOut.timeInterval = dataOut.ippSeconds * dataOut.nCohInt |
|
|
830 | 830 | dataOut.flagNoData = False |
|
|
831 | 831 | |
|
|
832 | 832 | #update Processing Header: |
|
|
833 | 833 | dataOut.processingHeaderObj.nCohInt = dataOut.nCohInt |
|
|
834 | 834 | |
|
|
835 | 835 | |
|
|
836 | 836 | return dataOut |
|
|
837 | 837 | |
|
|
838 | 838 | class Decoder(Operation): |
|
|
839 | 839 | |
|
|
840 | 840 | isConfig = False |
|
|
841 | 841 | __profIndex = 0 |
|
|
842 | 842 | |
|
|
843 | 843 | code = None |
|
|
844 | 844 | |
|
|
845 | 845 | nCode = None |
|
|
846 | 846 | nBaud = None |
|
|
847 | 847 | |
|
|
848 | 848 | def __init__(self, **kwargs): |
|
|
849 | 849 | |
|
|
850 | 850 | Operation.__init__(self, **kwargs) |
|
|
851 | 851 | |
|
|
852 | 852 | self.times = None |
|
|
853 | 853 | self.osamp = None |
|
|
854 | 854 | # self.__setValues = False |
|
|
855 | 855 | self.isConfig = False |
|
|
856 | 856 | self.setupReq = False |
|
|
857 | 857 | def setup(self, code, osamp, dataOut): |
|
|
858 | 858 | |
|
|
859 | 859 | self.__profIndex = 0 |
|
|
860 | 860 | |
|
|
861 | 861 | self.code = code |
|
|
862 | 862 | |
|
|
863 | 863 | self.nCode = len(code) |
|
|
864 | 864 | self.nBaud = len(code[0]) |
|
|
865 | 865 | if (osamp != None) and (osamp >1): |
|
|
866 | 866 | self.osamp = osamp |
|
|
867 | 867 | self.code = numpy.repeat(code, repeats=self.osamp, axis=1) |
|
|
868 | 868 | self.nBaud = self.nBaud*self.osamp |
|
|
869 | 869 | |
|
|
870 | 870 | self.__nChannels = dataOut.nChannels |
|
|
871 | 871 | self.__nProfiles = dataOut.nProfiles |
|
|
872 | 872 | self.__nHeis = dataOut.nHeights |
|
|
873 | 873 | |
|
|
874 | 874 | if self.__nHeis < self.nBaud: |
|
|
875 | 875 | raise ValueError('Number of heights (%d) should be greater than number of bauds (%d)' %(self.__nHeis, self.nBaud)) |
|
|
876 | 876 | |
|
|
877 | 877 | #Frequency |
|
|
878 | 878 | __codeBuffer = numpy.zeros((self.nCode, self.__nHeis), dtype=numpy.complex) |
|
|
879 | 879 | |
|
|
880 | 880 | __codeBuffer[:,0:self.nBaud] = self.code |
|
|
881 | 881 | |
|
|
882 | 882 | self.fft_code = numpy.conj(numpy.fft.fft(__codeBuffer, axis=1)) |
|
|
883 | 883 | |
|
|
884 | 884 | if dataOut.flagDataAsBlock: |
|
|
885 | 885 | |
|
|
886 | 886 | self.ndatadec = self.__nHeis #- self.nBaud + 1 |
|
|
887 | 887 | |
|
|
888 | 888 | self.datadecTime = numpy.zeros((self.__nChannels, self.__nProfiles, self.ndatadec), dtype=numpy.complex) |
|
|
889 | 889 | |
|
|
890 | 890 | else: |
|
|
891 | 891 | |
|
|
892 | 892 | #Time |
|
|
893 | 893 | self.ndatadec = self.__nHeis #- self.nBaud + 1 |
|
|
894 | 894 | |
|
|
895 | 895 | self.datadecTime = numpy.zeros((self.__nChannels, self.ndatadec), dtype=numpy.complex) |
|
|
896 | 896 | |
|
|
897 | 897 | def __convolutionInFreq(self, data): |
|
|
898 | 898 | |
|
|
899 | 899 | fft_code = self.fft_code[self.__profIndex].reshape(1,-1) |
|
|
900 | 900 | |
|
|
901 | 901 | fft_data = numpy.fft.fft(data, axis=1) |
|
|
902 | 902 | |
|
|
903 | 903 | conv = fft_data*fft_code |
|
|
904 | 904 | |
|
|
905 | 905 | data = numpy.fft.ifft(conv,axis=1) |
|
|
906 | 906 | |
|
|
907 | 907 | return data |
|
|
908 | 908 | |
|
|
909 | 909 | def __convolutionInFreqOpt(self, data): |
|
|
910 | 910 | |
|
|
911 | 911 | raise NotImplementedError |
|
|
912 | 912 | |
|
|
913 | 913 | def __convolutionInTime(self, data): |
|
|
914 | 914 | |
|
|
915 | 915 | code = self.code[self.__profIndex] |
|
|
916 | 916 | for i in range(self.__nChannels): |
|
|
917 | 917 | self.datadecTime[i,:] = numpy.correlate(data[i,:], code, mode='full')[self.nBaud-1:] |
|
|
918 | 918 | |
|
|
919 | 919 | return self.datadecTime |
|
|
920 | 920 | |
|
|
921 | 921 | def __convolutionByBlockInTime(self, data): |
|
|
922 | 922 | |
|
|
923 | 923 | repetitions = int(self.__nProfiles / self.nCode) |
|
|
924 | 924 | junk = numpy.lib.stride_tricks.as_strided(self.code, (repetitions, self.code.size), (0, self.code.itemsize)) |
|
|
925 | 925 | junk = junk.flatten() |
|
|
926 | 926 | code_block = numpy.reshape(junk, (self.nCode*repetitions, self.nBaud)) |
|
|
927 | 927 | profilesList = range(self.__nProfiles) |
|
|
928 | 928 | |
|
|
929 | 929 | for i in range(self.__nChannels): |
|
|
930 | 930 | for j in profilesList: |
|
|
931 | 931 | self.datadecTime[i,j,:] = numpy.correlate(data[i,j,:], code_block[j,:], mode='full')[self.nBaud-1:] |
|
|
932 | 932 | return self.datadecTime |
|
|
933 | 933 | |
|
|
934 | 934 | def __convolutionByBlockInFreq(self, data): |
|
|
935 | 935 | |
|
|
936 | 936 | raise NotImplementedError("Decoder by frequency fro Blocks not implemented") |
|
|
937 | 937 | |
|
|
938 | 938 | |
|
|
939 | 939 | fft_code = self.fft_code[self.__profIndex].reshape(1,-1) |
|
|
940 | 940 | |
|
|
941 | 941 | fft_data = numpy.fft.fft(data, axis=2) |
|
|
942 | 942 | |
|
|
943 | 943 | conv = fft_data*fft_code |
|
|
944 | 944 | |
|
|
945 | 945 | data = numpy.fft.ifft(conv,axis=2) |
|
|
946 | 946 | |
|
|
947 | 947 | return data |
|
|
948 | 948 | |
|
|
949 | 949 | |
|
|
950 | 950 | def run(self, dataOut, code=None, nCode=None, nBaud=None, mode = 0, osamp=None, times=None): |
|
|
951 | 951 | |
|
|
952 | 952 | if dataOut.flagDecodeData: |
|
|
953 | 953 | print("This data is already decoded, recoding again ...") |
|
|
954 | 954 | |
|
|
955 | 955 | if not self.isConfig: |
|
|
956 | 956 | |
|
|
957 | 957 | if code is None: |
|
|
958 | 958 | if dataOut.code is None: |
|
|
959 | 959 | raise ValueError("Code could not be read from %s instance. Enter a value in Code parameter" %dataOut.type) |
|
|
960 | 960 | |
|
|
961 | 961 | code = dataOut.code |
|
|
962 | 962 | else: |
|
|
963 | 963 | code = numpy.array(code).reshape(nCode,nBaud) |
|
|
964 | 964 | self.setup(code, osamp, dataOut) |
|
|
965 | 965 | |
|
|
966 | 966 | self.isConfig = True |
|
|
967 | 967 | |
|
|
968 | 968 | if mode == 3: |
|
|
969 | 969 | sys.stderr.write("Decoder Warning: mode=%d is not valid, using mode=0\n" %mode) |
|
|
970 | 970 | |
|
|
971 | 971 | if times != None: |
|
|
972 | 972 | sys.stderr.write("Decoder Warning: Argument 'times' in not used anymore\n") |
|
|
973 | 973 | |
|
|
974 | 974 | if self.code is None: |
|
|
975 | 975 | print("Fail decoding: Code is not defined.") |
|
|
976 | 976 | return |
|
|
977 | 977 | |
|
|
978 | 978 | self.__nProfiles = dataOut.nProfiles |
|
|
979 | 979 | datadec = None |
|
|
980 | 980 | |
|
|
981 | 981 | if mode == 3: |
|
|
982 | 982 | mode = 0 |
|
|
983 | 983 | |
|
|
984 | 984 | if dataOut.flagDataAsBlock: |
|
|
985 | 985 | """ |
|
|
986 | 986 | Decoding when data have been read as block, |
|
|
987 | 987 | """ |
|
|
988 | 988 | |
|
|
989 | 989 | if mode == 0: |
|
|
990 | 990 | datadec = self.__convolutionByBlockInTime(dataOut.data) |
|
|
991 | 991 | if mode == 1: |
|
|
992 | 992 | datadec = self.__convolutionByBlockInFreq(dataOut.data) |
|
|
993 | 993 | else: |
|
|
994 | 994 | """ |
|
|
995 | 995 | Decoding when data have been read profile by profile |
|
|
996 | 996 | """ |
|
|
997 | 997 | if mode == 0: |
|
|
998 | 998 | datadec = self.__convolutionInTime(dataOut.data) |
|
|
999 | 999 | |
|
|
1000 | 1000 | if mode == 1: |
|
|
1001 | 1001 | datadec = self.__convolutionInFreq(dataOut.data) |
|
|
1002 | 1002 | |
|
|
1003 | 1003 | if mode == 2: |
|
|
1004 | 1004 | datadec = self.__convolutionInFreqOpt(dataOut.data) |
|
|
1005 | 1005 | |
|
|
1006 | 1006 | if datadec is None: |
|
|
1007 | 1007 | raise ValueError("Codification mode selected is not valid: mode=%d. Try selecting 0 or 1" %mode) |
|
|
1008 | 1008 | |
|
|
1009 | 1009 | dataOut.code = self.code |
|
|
1010 | 1010 | dataOut.nCode = self.nCode |
|
|
1011 | 1011 | dataOut.nBaud = self.nBaud |
|
|
1012 | 1012 | |
|
|
1013 | 1013 | dataOut.data = datadec |
|
|
1014 | 1014 | dataOut.heightList = dataOut.heightList[0:datadec.shape[-1]] |
|
|
1015 | 1015 | dataOut.flagDecodeData = True #asumo q la data esta decodificada |
|
|
1016 | 1016 | |
|
|
1017 | 1017 | |
|
|
1018 | 1018 | #update Processing Header: |
|
|
1019 | 1019 | dataOut.radarControllerHeaderObj.code = self.code |
|
|
1020 | 1020 | dataOut.radarControllerHeaderObj.nCode = self.nCode |
|
|
1021 | 1021 | dataOut.radarControllerHeaderObj.nBaud = self.nBaud |
|
|
1022 | 1022 | dataOut.radarControllerHeaderObj.nOsamp = osamp |
|
|
1023 | 1023 | #update Processing Header: |
|
|
1024 | 1024 | dataOut.processingHeaderObj.heightList = dataOut.heightList |
|
|
1025 | 1025 | dataOut.processingHeaderObj.heightResolution = dataOut.heightList[1]-dataOut.heightList[0] |
|
|
1026 | 1026 | |
|
|
1027 | 1027 | if self.__profIndex == self.nCode-1: |
|
|
1028 | 1028 | self.__profIndex = 0 |
|
|
1029 | 1029 | return dataOut |
|
|
1030 | 1030 | |
|
|
1031 | 1031 | self.__profIndex += 1 |
|
|
1032 | 1032 | |
|
|
1033 | 1033 | return dataOut |
|
|
1034 | 1034 | |
|
|
1035 | 1035 | class ProfileConcat(Operation): |
|
|
1036 | 1036 | |
|
|
1037 | 1037 | isConfig = False |
|
|
1038 | 1038 | buffer = None |
|
|
1039 | 1039 | |
|
|
1040 | 1040 | def __init__(self, **kwargs): |
|
|
1041 | 1041 | |
|
|
1042 | 1042 | Operation.__init__(self, **kwargs) |
|
|
1043 | 1043 | self.profileIndex = 0 |
|
|
1044 | 1044 | |
|
|
1045 | 1045 | def reset(self): |
|
|
1046 | 1046 | self.buffer = numpy.zeros_like(self.buffer) |
|
|
1047 | 1047 | self.start_index = 0 |
|
|
1048 | 1048 | self.times = 1 |
|
|
1049 | 1049 | |
|
|
1050 | 1050 | def setup(self, data, m, n=1): |
|
|
1051 | 1051 | self.buffer = numpy.zeros((data.shape[0],data.shape[1]*m),dtype=type(data[0,0])) |
|
|
1052 | 1052 | self.nHeights = data.shape[1]#.nHeights |
|
|
1053 | 1053 | self.start_index = 0 |
|
|
1054 | 1054 | self.times = 1 |
|
|
1055 | 1055 | |
|
|
1056 | 1056 | def concat(self, data): |
|
|
1057 | 1057 | |
|
|
1058 | 1058 | self.buffer[:,self.start_index:self.nHeights*self.times] = data.copy() |
|
|
1059 | 1059 | self.start_index = self.start_index + self.nHeights |
|
|
1060 | 1060 | |
|
|
1061 | 1061 | def run(self, dataOut, m): |
|
|
1062 | 1062 | dataOut.flagNoData = True |
|
|
1063 | 1063 | |
|
|
1064 | 1064 | if not self.isConfig: |
|
|
1065 | 1065 | self.setup(dataOut.data, m, 1) |
|
|
1066 | 1066 | self.isConfig = True |
|
|
1067 | 1067 | |
|
|
1068 | 1068 | if dataOut.flagDataAsBlock: |
|
|
1069 | 1069 | raise ValueError("ProfileConcat can only be used when voltage have been read profile by profile, getBlock = False") |
|
|
1070 | 1070 | |
|
|
1071 | 1071 | else: |
|
|
1072 | 1072 | self.concat(dataOut.data) |
|
|
1073 | 1073 | self.times += 1 |
|
|
1074 | 1074 | if self.times > m: |
|
|
1075 | 1075 | dataOut.data = self.buffer |
|
|
1076 | 1076 | self.reset() |
|
|
1077 | 1077 | dataOut.flagNoData = False |
|
|
1078 | 1078 | # se deben actualizar mas propiedades del header y del objeto dataOut, por ejemplo, las alturas |
|
|
1079 | 1079 | deltaHeight = dataOut.heightList[1] - dataOut.heightList[0] |
|
|
1080 | 1080 | xf = dataOut.heightList[0] + dataOut.nHeights * deltaHeight * m |
|
|
1081 | 1081 | dataOut.heightList = numpy.arange(dataOut.heightList[0], xf, deltaHeight) |
|
|
1082 | 1082 | dataOut.ippSeconds *= m |
|
|
1083 | 1083 | |
|
|
1084 | 1084 | #update Processing Header: |
|
|
1085 | 1085 | dataOut.processingHeaderObj.heightList = dataOut.heightList |
|
|
1086 | 1086 | dataOut.processingHeaderObj.ipp = dataOut.ippSeconds |
|
|
1087 | 1087 | |
|
|
1088 | 1088 | return dataOut |
|
|
1089 | 1089 | |
|
|
1090 | 1090 | class ProfileSelector(Operation): |
|
|
1091 | 1091 | |
|
|
1092 | 1092 | profileIndex = None |
|
|
1093 | 1093 | # Tamanho total de los perfiles |
|
|
1094 | 1094 | nProfiles = None |
|
|
1095 | 1095 | |
|
|
1096 | 1096 | def __init__(self, **kwargs): |
|
|
1097 | 1097 | |
|
|
1098 | 1098 | Operation.__init__(self, **kwargs) |
|
|
1099 | 1099 | self.profileIndex = 0 |
|
|
1100 | 1100 | |
|
|
1101 | 1101 | def incProfileIndex(self): |
|
|
1102 | 1102 | |
|
|
1103 | 1103 | self.profileIndex += 1 |
|
|
1104 | 1104 | |
|
|
1105 | 1105 | if self.profileIndex >= self.nProfiles: |
|
|
1106 | 1106 | self.profileIndex = 0 |
|
|
1107 | 1107 | |
|
|
1108 | 1108 | def isThisProfileInRange(self, profileIndex, minIndex, maxIndex): |
|
|
1109 | 1109 | |
|
|
1110 | 1110 | if profileIndex < minIndex: |
|
|
1111 | 1111 | return False |
|
|
1112 | 1112 | |
|
|
1113 | 1113 | if profileIndex > maxIndex: |
|
|
1114 | 1114 | return False |
|
|
1115 | 1115 | |
|
|
1116 | 1116 | return True |
|
|
1117 | 1117 | |
|
|
1118 | 1118 | def isThisProfileInList(self, profileIndex, profileList): |
|
|
1119 | 1119 | |
|
|
1120 | 1120 | if profileIndex not in profileList: |
|
|
1121 | 1121 | return False |
|
|
1122 | 1122 | |
|
|
1123 | 1123 | return True |
|
|
1124 | 1124 | |
|
|
1125 | 1125 | def run(self, dataOut, profileList=None, profileRangeList=None, beam=None, byblock=False, rangeList = None, nProfiles=None): |
|
|
1126 | 1126 | |
|
|
1127 | 1127 | """ |
|
|
1128 | 1128 | ProfileSelector: |
|
|
1129 | 1129 | |
|
|
1130 | 1130 | Inputs: |
|
|
1131 | 1131 | profileList : Index of profiles selected. Example: profileList = (0,1,2,7,8) |
|
|
1132 | 1132 | |
|
|
1133 | 1133 | profileRangeList : Minimum and maximum profile indexes. Example: profileRangeList = (4, 30) |
|
|
1134 | 1134 | |
|
|
1135 | 1135 | rangeList : List of profile ranges. Example: rangeList = ((4, 30), (32, 64), (128, 256)) |
|
|
1136 | 1136 | |
|
|
1137 | 1137 | """ |
|
|
1138 | 1138 | |
|
|
1139 | 1139 | if rangeList is not None: |
|
|
1140 | 1140 | if type(rangeList[0]) not in (tuple, list): |
|
|
1141 | 1141 | rangeList = [rangeList] |
|
|
1142 | 1142 | |
|
|
1143 | 1143 | dataOut.flagNoData = True |
|
|
1144 | 1144 | |
|
|
1145 | 1145 | if dataOut.flagDataAsBlock: |
|
|
1146 | 1146 | """ |
|
|
1147 | 1147 | data dimension = [nChannels, nProfiles, nHeis] |
|
|
1148 | 1148 | """ |
|
|
1149 | 1149 | if profileList != None: |
|
|
1150 | 1150 | dataOut.data = dataOut.data[:,profileList,:] |
|
|
1151 | 1151 | |
|
|
1152 | 1152 | if profileRangeList != None: |
|
|
1153 | 1153 | minIndex = profileRangeList[0] |
|
|
1154 | 1154 | maxIndex = profileRangeList[1] |
|
|
1155 | 1155 | profileList = list(range(minIndex, maxIndex+1)) |
|
|
1156 | 1156 | |
|
|
1157 | 1157 | dataOut.data = dataOut.data[:,minIndex:maxIndex+1,:] |
|
|
1158 | 1158 | |
|
|
1159 | 1159 | if rangeList != None: |
|
|
1160 | 1160 | |
|
|
1161 | 1161 | profileList = [] |
|
|
1162 | 1162 | |
|
|
1163 | 1163 | for thisRange in rangeList: |
|
|
1164 | 1164 | minIndex = thisRange[0] |
|
|
1165 | 1165 | maxIndex = thisRange[1] |
|
|
1166 | 1166 | |
|
|
1167 | 1167 | profileList.extend(list(range(minIndex, maxIndex+1))) |
|
|
1168 | 1168 | |
|
|
1169 | 1169 | dataOut.data = dataOut.data[:,profileList,:] |
|
|
1170 | 1170 | |
|
|
1171 | 1171 | dataOut.nProfiles = len(profileList) |
|
|
1172 | 1172 | dataOut.profileIndex = dataOut.nProfiles - 1 |
|
|
1173 | 1173 | dataOut.flagNoData = False |
|
|
1174 | 1174 | |
|
|
1175 | 1175 | return dataOut |
|
|
1176 | 1176 | |
|
|
1177 | 1177 | """ |
|
|
1178 | 1178 | data dimension = [nChannels, nHeis] |
|
|
1179 | 1179 | """ |
|
|
1180 | 1180 | |
|
|
1181 | 1181 | if profileList != None: |
|
|
1182 | 1182 | |
|
|
1183 | 1183 | if self.isThisProfileInList(dataOut.profileIndex, profileList): |
|
|
1184 | 1184 | |
|
|
1185 | 1185 | self.nProfiles = len(profileList) |
|
|
1186 | 1186 | dataOut.nProfiles = self.nProfiles |
|
|
1187 | 1187 | dataOut.profileIndex = self.profileIndex |
|
|
1188 | 1188 | dataOut.flagNoData = False |
|
|
1189 | 1189 | |
|
|
1190 | 1190 | self.incProfileIndex() |
|
|
1191 | 1191 | return dataOut |
|
|
1192 | 1192 | |
|
|
1193 | 1193 | if profileRangeList != None: |
|
|
1194 | 1194 | |
|
|
1195 | 1195 | minIndex = profileRangeList[0] |
|
|
1196 | 1196 | maxIndex = profileRangeList[1] |
|
|
1197 | 1197 | |
|
|
1198 | 1198 | if self.isThisProfileInRange(dataOut.profileIndex, minIndex, maxIndex): |
|
|
1199 | 1199 | |
|
|
1200 | 1200 | self.nProfiles = maxIndex - minIndex + 1 |
|
|
1201 | 1201 | dataOut.nProfiles = self.nProfiles |
|
|
1202 | 1202 | dataOut.profileIndex = self.profileIndex |
|
|
1203 | 1203 | dataOut.flagNoData = False |
|
|
1204 | 1204 | |
|
|
1205 | 1205 | self.incProfileIndex() |
|
|
1206 | 1206 | return dataOut |
|
|
1207 | 1207 | |
|
|
1208 | 1208 | if rangeList != None: |
|
|
1209 | 1209 | |
|
|
1210 | 1210 | nProfiles = 0 |
|
|
1211 | 1211 | |
|
|
1212 | 1212 | for thisRange in rangeList: |
|
|
1213 | 1213 | minIndex = thisRange[0] |
|
|
1214 | 1214 | maxIndex = thisRange[1] |
|
|
1215 | 1215 | |
|
|
1216 | 1216 | nProfiles += maxIndex - minIndex + 1 |
|
|
1217 | 1217 | |
|
|
1218 | 1218 | for thisRange in rangeList: |
|
|
1219 | 1219 | |
|
|
1220 | 1220 | minIndex = thisRange[0] |
|
|
1221 | 1221 | maxIndex = thisRange[1] |
|
|
1222 | 1222 | |
|
|
1223 | 1223 | if self.isThisProfileInRange(dataOut.profileIndex, minIndex, maxIndex): |
|
|
1224 | 1224 | |
|
|
1225 | 1225 | self.nProfiles = nProfiles |
|
|
1226 | 1226 | dataOut.nProfiles = self.nProfiles |
|
|
1227 | 1227 | dataOut.profileIndex = self.profileIndex |
|
|
1228 | 1228 | dataOut.flagNoData = False |
|
|
1229 | 1229 | |
|
|
1230 | 1230 | self.incProfileIndex() |
|
|
1231 | 1231 | |
|
|
1232 | 1232 | break |
|
|
1233 | 1233 | |
|
|
1234 | 1234 | return dataOut |
|
|
1235 | 1235 | |
|
|
1236 | 1236 | |
|
|
1237 | 1237 | if beam != None: #beam is only for AMISR data |
|
|
1238 | 1238 | if self.isThisProfileInList(dataOut.profileIndex, dataOut.beamRangeDict[beam]): |
|
|
1239 | 1239 | dataOut.flagNoData = False |
|
|
1240 | 1240 | dataOut.profileIndex = self.profileIndex |
|
|
1241 | 1241 | |
|
|
1242 | 1242 | self.incProfileIndex() |
|
|
1243 | 1243 | |
|
|
1244 | 1244 | return dataOut |
|
|
1245 | 1245 | |
|
|
1246 | 1246 | raise ValueError("ProfileSelector needs profileList, profileRangeList or rangeList parameter") |
|
|
1247 | 1247 | |
|
|
1248 | 1248 | |
|
|
1249 | 1249 | class Reshaper(Operation): |
|
|
1250 | 1250 | |
|
|
1251 | 1251 | def __init__(self, **kwargs): |
|
|
1252 | 1252 | |
|
|
1253 | 1253 | Operation.__init__(self, **kwargs) |
|
|
1254 | 1254 | |
|
|
1255 | 1255 | self.__buffer = None |
|
|
1256 | 1256 | self.__nitems = 0 |
|
|
1257 | 1257 | |
|
|
1258 | 1258 | def __appendProfile(self, dataOut, nTxs): |
|
|
1259 | 1259 | |
|
|
1260 | 1260 | if self.__buffer is None: |
|
|
1261 | 1261 | shape = (dataOut.nChannels, int(dataOut.nHeights/nTxs) ) |
|
|
1262 | 1262 | self.__buffer = numpy.empty(shape, dtype = dataOut.data.dtype) |
|
|
1263 | 1263 | |
|
|
1264 | 1264 | ini = dataOut.nHeights * self.__nitems |
|
|
1265 | 1265 | end = ini + dataOut.nHeights |
|
|
1266 | 1266 | |
|
|
1267 | 1267 | self.__buffer[:, ini:end] = dataOut.data |
|
|
1268 | 1268 | |
|
|
1269 | 1269 | self.__nitems += 1 |
|
|
1270 | 1270 | |
|
|
1271 | 1271 | return int(self.__nitems*nTxs) |
|
|
1272 | 1272 | |
|
|
1273 | 1273 | def __getBuffer(self): |
|
|
1274 | 1274 | |
|
|
1275 | 1275 | if self.__nitems == int(1./self.__nTxs): |
|
|
1276 | 1276 | |
|
|
1277 | 1277 | self.__nitems = 0 |
|
|
1278 | 1278 | |
|
|
1279 | 1279 | return self.__buffer.copy() |
|
|
1280 | 1280 | |
|
|
1281 | 1281 | return None |
|
|
1282 | 1282 | |
|
|
1283 | 1283 | def __checkInputs(self, dataOut, shape, nTxs): |
|
|
1284 | 1284 | |
|
|
1285 | 1285 | if shape is None and nTxs is None: |
|
|
1286 | 1286 | raise ValueError("Reshaper: shape of factor should be defined") |
|
|
1287 | 1287 | |
|
|
1288 | 1288 | if nTxs: |
|
|
1289 | 1289 | if nTxs < 0: |
|
|
1290 | 1290 | raise ValueError("nTxs should be greater than 0") |
|
|
1291 | 1291 | |
|
|
1292 | 1292 | if nTxs < 1 and dataOut.nProfiles % (1./nTxs) != 0: |
|
|
1293 | 1293 | raise ValueError("nProfiles= %d is not divisibled by (1./nTxs) = %f" %(dataOut.nProfiles, (1./nTxs))) |
|
|
1294 | 1294 | |
|
|
1295 | 1295 | shape = [dataOut.nChannels, dataOut.nProfiles*nTxs, dataOut.nHeights/nTxs] |
|
|
1296 | 1296 | |
|
|
1297 | 1297 | return shape, nTxs |
|
|
1298 | 1298 | |
|
|
1299 | 1299 | if len(shape) != 2 and len(shape) != 3: |
|
|
1300 | 1300 | raise ValueError("shape dimension should be equal to 2 or 3. shape = (nProfiles, nHeis) or (nChannels, nProfiles, nHeis). Actually shape = (%d, %d, %d)" %(dataOut.nChannels, dataOut.nProfiles, dataOut.nHeights)) |
|
|
1301 | 1301 | |
|
|
1302 | 1302 | if len(shape) == 2: |
|
|
1303 | 1303 | shape_tuple = [dataOut.nChannels] |
|
|
1304 | 1304 | shape_tuple.extend(shape) |
|
|
1305 | 1305 | else: |
|
|
1306 | 1306 | shape_tuple = list(shape) |
|
|
1307 | 1307 | |
|
|
1308 | 1308 | nTxs = 1.0*shape_tuple[1]/dataOut.nProfiles |
|
|
1309 | 1309 | |
|
|
1310 | 1310 | return shape_tuple, nTxs |
|
|
1311 | 1311 | |
|
|
1312 | 1312 | def run(self, dataOut, shape=None, nTxs=None): |
|
|
1313 | 1313 | |
|
|
1314 | 1314 | shape_tuple, self.__nTxs = self.__checkInputs(dataOut, shape, nTxs) |
|
|
1315 | 1315 | |
|
|
1316 | 1316 | dataOut.flagNoData = True |
|
|
1317 | 1317 | profileIndex = None |
|
|
1318 | 1318 | |
|
|
1319 | 1319 | if dataOut.flagDataAsBlock: |
|
|
1320 | 1320 | |
|
|
1321 | 1321 | dataOut.data = numpy.reshape(dataOut.data, shape_tuple) |
|
|
1322 | 1322 | dataOut.flagNoData = False |
|
|
1323 | 1323 | |
|
|
1324 | 1324 | profileIndex = int(dataOut.nProfiles*self.__nTxs) - 1 |
|
|
1325 | 1325 | |
|
|
1326 | 1326 | else: |
|
|
1327 | 1327 | |
|
|
1328 | 1328 | if self.__nTxs < 1: |
|
|
1329 | 1329 | |
|
|
1330 | 1330 | self.__appendProfile(dataOut, self.__nTxs) |
|
|
1331 | 1331 | new_data = self.__getBuffer() |
|
|
1332 | 1332 | |
|
|
1333 | 1333 | if new_data is not None: |
|
|
1334 | 1334 | dataOut.data = new_data |
|
|
1335 | 1335 | dataOut.flagNoData = False |
|
|
1336 | 1336 | |
|
|
1337 | 1337 | profileIndex = dataOut.profileIndex*nTxs |
|
|
1338 | 1338 | |
|
|
1339 | 1339 | else: |
|
|
1340 | 1340 | raise ValueError("nTxs should be greater than 0 and lower than 1, or use VoltageReader(..., getblock=True)") |
|
|
1341 | 1341 | |
|
|
1342 | 1342 | deltaHeight = dataOut.heightList[1] - dataOut.heightList[0] |
|
|
1343 | 1343 | |
|
|
1344 | 1344 | dataOut.heightList = numpy.arange(dataOut.nHeights/self.__nTxs) * deltaHeight + dataOut.heightList[0] |
|
|
1345 | 1345 | |
|
|
1346 | 1346 | dataOut.nProfiles = int(dataOut.nProfiles*self.__nTxs) |
|
|
1347 | 1347 | |
|
|
1348 | 1348 | dataOut.profileIndex = profileIndex |
|
|
1349 | 1349 | |
|
|
1350 | 1350 | dataOut.ippSeconds /= self.__nTxs |
|
|
1351 | 1351 | |
|
|
1352 | 1352 | return dataOut |
|
|
1353 | 1353 | |
|
|
1354 | 1354 | class SplitProfiles(Operation): |
|
|
1355 | 1355 | |
|
|
1356 | 1356 | def __init__(self, **kwargs): |
|
|
1357 | 1357 | |
|
|
1358 | 1358 | Operation.__init__(self, **kwargs) |
|
|
1359 | 1359 | |
|
|
1360 | 1360 | def run(self, dataOut, n): |
|
|
1361 | 1361 | |
|
|
1362 | 1362 | dataOut.flagNoData = True |
|
|
1363 | 1363 | profileIndex = None |
|
|
1364 | 1364 | |
|
|
1365 | 1365 | if dataOut.flagDataAsBlock: |
|
|
1366 | 1366 | |
|
|
1367 | 1367 | #nchannels, nprofiles, nsamples |
|
|
1368 | 1368 | shape = dataOut.data.shape |
|
|
1369 | 1369 | |
|
|
1370 | 1370 | if shape[2] % n != 0: |
|
|
1371 | 1371 | raise ValueError("Could not split the data, n=%d has to be multiple of %d" %(n, shape[2])) |
|
|
1372 | 1372 | |
|
|
1373 | 1373 | new_shape = shape[0], shape[1]*n, int(shape[2]/n) |
|
|
1374 | 1374 | |
|
|
1375 | 1375 | dataOut.data = numpy.reshape(dataOut.data, new_shape) |
|
|
1376 | 1376 | dataOut.flagNoData = False |
|
|
1377 | 1377 | |
|
|
1378 | 1378 | profileIndex = int(dataOut.nProfiles/n) - 1 |
|
|
1379 | 1379 | |
|
|
1380 | 1380 | else: |
|
|
1381 | 1381 | |
|
|
1382 | 1382 | raise ValueError("Could not split the data when is read Profile by Profile. Use VoltageReader(..., getblock=True)") |
|
|
1383 | 1383 | |
|
|
1384 | 1384 | deltaHeight = dataOut.heightList[1] - dataOut.heightList[0] |
|
|
1385 | 1385 | |
|
|
1386 | 1386 | dataOut.heightList = numpy.arange(dataOut.nHeights/n) * deltaHeight + dataOut.heightList[0] |
|
|
1387 | 1387 | |
|
|
1388 | 1388 | dataOut.nProfiles = int(dataOut.nProfiles*n) |
|
|
1389 | 1389 | |
|
|
1390 | 1390 | dataOut.profileIndex = profileIndex |
|
|
1391 | 1391 | |
|
|
1392 | 1392 | dataOut.ippSeconds /= n |
|
|
1393 | 1393 | |
|
|
1394 | 1394 | return dataOut |
|
|
1395 | 1395 | |
|
|
1396 | 1396 | class CombineProfiles(Operation): |
|
|
1397 | 1397 | def __init__(self, **kwargs): |
|
|
1398 | 1398 | |
|
|
1399 | 1399 | Operation.__init__(self, **kwargs) |
|
|
1400 | 1400 | |
|
|
1401 | 1401 | self.__remData = None |
|
|
1402 | 1402 | self.__profileIndex = 0 |
|
|
1403 | 1403 | |
|
|
1404 | 1404 | def run(self, dataOut, n): |
|
|
1405 | 1405 | |
|
|
1406 | 1406 | dataOut.flagNoData = True |
|
|
1407 | 1407 | profileIndex = None |
|
|
1408 | 1408 | |
|
|
1409 | 1409 | if dataOut.flagDataAsBlock: |
|
|
1410 | 1410 | |
|
|
1411 | 1411 | #nchannels, nprofiles, nsamples |
|
|
1412 | 1412 | shape = dataOut.data.shape |
|
|
1413 | 1413 | new_shape = shape[0], shape[1]/n, shape[2]*n |
|
|
1414 | 1414 | |
|
|
1415 | 1415 | if shape[1] % n != 0: |
|
|
1416 | 1416 | raise ValueError("Could not split the data, n=%d has to be multiple of %d" %(n, shape[1])) |
|
|
1417 | 1417 | |
|
|
1418 | 1418 | dataOut.data = numpy.reshape(dataOut.data, new_shape) |
|
|
1419 | 1419 | dataOut.flagNoData = False |
|
|
1420 | 1420 | |
|
|
1421 | 1421 | profileIndex = int(dataOut.nProfiles*n) - 1 |
|
|
1422 | 1422 | |
|
|
1423 | 1423 | else: |
|
|
1424 | 1424 | |
|
|
1425 | 1425 | #nchannels, nsamples |
|
|
1426 | 1426 | if self.__remData is None: |
|
|
1427 | 1427 | newData = dataOut.data |
|
|
1428 | 1428 | else: |
|
|
1429 | 1429 | newData = numpy.concatenate((self.__remData, dataOut.data), axis=1) |
|
|
1430 | 1430 | |
|
|
1431 | 1431 | self.__profileIndex += 1 |
|
|
1432 | 1432 | |
|
|
1433 | 1433 | if self.__profileIndex < n: |
|
|
1434 | 1434 | self.__remData = newData |
|
|
1435 | 1435 | #continue |
|
|
1436 | 1436 | return |
|
|
1437 | 1437 | |
|
|
1438 | 1438 | self.__profileIndex = 0 |
|
|
1439 | 1439 | self.__remData = None |
|
|
1440 | 1440 | |
|
|
1441 | 1441 | dataOut.data = newData |
|
|
1442 | 1442 | dataOut.flagNoData = False |
|
|
1443 | 1443 | |
|
|
1444 | 1444 | profileIndex = dataOut.profileIndex/n |
|
|
1445 | 1445 | |
|
|
1446 | 1446 | |
|
|
1447 | 1447 | deltaHeight = dataOut.heightList[1] - dataOut.heightList[0] |
|
|
1448 | 1448 | |
|
|
1449 | 1449 | dataOut.heightList = numpy.arange(dataOut.nHeights*n) * deltaHeight + dataOut.heightList[0] |
|
|
1450 | 1450 | |
|
|
1451 | 1451 | dataOut.nProfiles = int(dataOut.nProfiles/n) |
|
|
1452 | 1452 | |
|
|
1453 | 1453 | dataOut.profileIndex = profileIndex |
|
|
1454 | 1454 | |
|
|
1455 | 1455 | dataOut.ippSeconds *= n |
|
|
1456 | 1456 | |
|
|
1457 | 1457 | return dataOut |
|
|
1458 | 1458 | |
|
|
1459 | 1459 | class PulsePairVoltage(Operation): |
|
|
1460 | 1460 | ''' |
|
|
1461 | 1461 | Function PulsePair(Signal Power, Velocity) |
|
|
1462 | 1462 | The real component of Lag[0] provides Intensity Information |
|
|
1463 | 1463 | The imag component of Lag[1] Phase provides Velocity Information |
|
|
1464 | 1464 | |
|
|
1465 | 1465 | Configuration Parameters: |
|
|
1466 | 1466 | nPRF = Number of Several PRF |
|
|
1467 | 1467 | theta = Degree Azimuth angel Boundaries |
|
|
1468 | 1468 | |
|
|
1469 | 1469 | Input: |
|
|
1470 | 1470 | self.dataOut |
|
|
1471 | 1471 | lag[N] |
|
|
1472 | 1472 | Affected: |
|
|
1473 | 1473 | self.dataOut.spc |
|
|
1474 | 1474 | ''' |
|
|
1475 | 1475 | isConfig = False |
|
|
1476 | 1476 | __profIndex = 0 |
|
|
1477 | 1477 | __initime = None |
|
|
1478 | 1478 | __lastdatatime = None |
|
|
1479 | 1479 | __buffer = None |
|
|
1480 | 1480 | noise = None |
|
|
1481 | 1481 | __dataReady = False |
|
|
1482 | 1482 | n = None |
|
|
1483 | 1483 | __nch = 0 |
|
|
1484 | 1484 | __nHeis = 0 |
|
|
1485 | 1485 | removeDC = False |
|
|
1486 | 1486 | ipp = None |
|
|
1487 | 1487 | lambda_ = 0 |
|
|
1488 | 1488 | |
|
|
1489 | 1489 | def __init__(self,**kwargs): |
|
|
1490 | 1490 | Operation.__init__(self,**kwargs) |
|
|
1491 | 1491 | |
|
|
1492 | 1492 | def setup(self, dataOut, n = None, removeDC=False): |
|
|
1493 | 1493 | ''' |
|
|
1494 | 1494 | n= Numero de PRF's de entrada |
|
|
1495 | 1495 | ''' |
|
|
1496 | 1496 | self.__initime = None |
|
|
1497 | 1497 | self.__lastdatatime = 0 |
|
|
1498 | 1498 | self.__dataReady = False |
|
|
1499 | 1499 | self.__buffer = 0 |
|
|
1500 | 1500 | self.__profIndex = 0 |
|
|
1501 | 1501 | self.noise = None |
|
|
1502 | 1502 | self.__nch = dataOut.nChannels |
|
|
1503 | 1503 | self.__nHeis = dataOut.nHeights |
|
|
1504 | 1504 | self.removeDC = removeDC |
|
|
1505 | 1505 | self.lambda_ = 3.0e8/(9345.0e6) |
|
|
1506 | 1506 | self.ippSec = dataOut.ippSeconds |
|
|
1507 | 1507 | self.nCohInt = dataOut.nCohInt |
|
|
1508 | 1508 | |
|
|
1509 | 1509 | if n == None: |
|
|
1510 | 1510 | raise ValueError("n should be specified.") |
|
|
1511 | 1511 | |
|
|
1512 | 1512 | if n != None: |
|
|
1513 | 1513 | if n<2: |
|
|
1514 | 1514 | raise ValueError("n should be greater than 2") |
|
|
1515 | 1515 | |
|
|
1516 | 1516 | self.n = n |
|
|
1517 | 1517 | self.__nProf = n |
|
|
1518 | 1518 | |
|
|
1519 | 1519 | self.__buffer = numpy.zeros((dataOut.nChannels, |
|
|
1520 | 1520 | n, |
|
|
1521 | 1521 | dataOut.nHeights), |
|
|
1522 | 1522 | dtype='complex') |
|
|
1523 | 1523 | |
|
|
1524 | 1524 | def putData(self,data): |
|
|
1525 | 1525 | ''' |
|
|
1526 | 1526 | Add a profile to he __buffer and increase in one the __profiel Index |
|
|
1527 | 1527 | ''' |
|
|
1528 | 1528 | self.__buffer[:,self.__profIndex,:]= data |
|
|
1529 | 1529 | self.__profIndex += 1 |
|
|
1530 | 1530 | return |
|
|
1531 | 1531 | |
|
|
1532 | 1532 | def pushData(self,dataOut): |
|
|
1533 | 1533 | ''' |
|
|
1534 | 1534 | Return the PULSEPAIR and the profiles used in the operation |
|
|
1535 | 1535 | Affected : self.__profileIndex |
|
|
1536 | 1536 | ''' |
|
|
1537 | 1537 | #----------------- Remove DC----------------------------------- |
|
|
1538 | 1538 | if self.removeDC==True: |
|
|
1539 | 1539 | mean = numpy.mean(self.__buffer,1) |
|
|
1540 | 1540 | tmp = mean.reshape(self.__nch,1,self.__nHeis) |
|
|
1541 | 1541 | dc= numpy.tile(tmp,[1,self.__nProf,1]) |
|
|
1542 | 1542 | self.__buffer = self.__buffer - dc |
|
|
1543 | 1543 | #------------------Calculo de Potencia ------------------------ |
|
|
1544 | 1544 | pair0 = self.__buffer*numpy.conj(self.__buffer) |
|
|
1545 | 1545 | pair0 = pair0.real |
|
|
1546 | 1546 | lag_0 = numpy.sum(pair0,1) |
|
|
1547 | 1547 | #------------------Calculo de Ruido x canal-------------------- |
|
|
1548 | 1548 | self.noise = numpy.zeros(self.__nch) |
|
|
1549 | 1549 | for i in range(self.__nch): |
|
|
1550 | 1550 | daux = numpy.sort(pair0[i,:,:],axis= None) |
|
|
1551 | 1551 | self.noise[i]=hildebrand_sekhon( daux ,self.nCohInt) |
|
|
1552 | 1552 | |
|
|
1553 | 1553 | self.noise = self.noise.reshape(self.__nch,1) |
|
|
1554 | 1554 | self.noise = numpy.tile(self.noise,[1,self.__nHeis]) |
|
|
1555 | 1555 | noise_buffer = self.noise.reshape(self.__nch,1,self.__nHeis) |
|
|
1556 | 1556 | noise_buffer = numpy.tile(noise_buffer,[1,self.__nProf,1]) |
|
|
1557 | 1557 | #------------------ Potencia recibida= P , Potencia senal = S , Ruido= N-- |
|
|
1558 | 1558 | #------------------ P= S+N ,P=lag_0/N --------------------------------- |
|
|
1559 | 1559 | #-------------------- Power -------------------------------------------------- |
|
|
1560 | 1560 | data_power = lag_0/(self.n*self.nCohInt) |
|
|
1561 | 1561 | #------------------ Senal --------------------------------------------------- |
|
|
1562 | 1562 | data_intensity = pair0 - noise_buffer |
|
|
1563 | 1563 | data_intensity = numpy.sum(data_intensity,axis=1)*(self.n*self.nCohInt)#*self.nCohInt) |
|
|
1564 | 1564 | #data_intensity = (lag_0-self.noise*self.n)*(self.n*self.nCohInt) |
|
|
1565 | 1565 | for i in range(self.__nch): |
|
|
1566 | 1566 | for j in range(self.__nHeis): |
|
|
1567 | 1567 | if data_intensity[i][j] < 0: |
|
|
1568 | 1568 | data_intensity[i][j] = numpy.min(numpy.absolute(data_intensity[i][j])) |
|
|
1569 | 1569 | |
|
|
1570 | 1570 | #----------------- Calculo de Frecuencia y Velocidad doppler-------- |
|
|
1571 | 1571 | pair1 = self.__buffer[:,:-1,:]*numpy.conjugate(self.__buffer[:,1:,:]) |
|
|
1572 | 1572 | lag_1 = numpy.sum(pair1,1) |
|
|
1573 | 1573 | data_freq = (-1/(2.0*math.pi*self.ippSec*self.nCohInt))*numpy.angle(lag_1) |
|
|
1574 | 1574 | data_velocity = (self.lambda_/2.0)*data_freq |
|
|
1575 | 1575 | |
|
|
1576 | 1576 | #---------------- Potencia promedio estimada de la Senal----------- |
|
|
1577 | 1577 | lag_0 = lag_0/self.n |
|
|
1578 | 1578 | S = lag_0-self.noise |
|
|
1579 | 1579 | |
|
|
1580 | 1580 | #---------------- Frecuencia Doppler promedio --------------------- |
|
|
1581 | 1581 | lag_1 = lag_1/(self.n-1) |
|
|
1582 | 1582 | R1 = numpy.abs(lag_1) |
|
|
1583 | 1583 | |
|
|
1584 | 1584 | #---------------- Calculo del SNR---------------------------------- |
|
|
1585 | 1585 | data_snrPP = S/self.noise |
|
|
1586 | 1586 | for i in range(self.__nch): |
|
|
1587 | 1587 | for j in range(self.__nHeis): |
|
|
1588 | 1588 | if data_snrPP[i][j] < 1.e-20: |
|
|
1589 | 1589 | data_snrPP[i][j] = 1.e-20 |
|
|
1590 | 1590 | |
|
|
1591 | 1591 | #----------------- Calculo del ancho espectral ---------------------- |
|
|
1592 | 1592 | L = S/R1 |
|
|
1593 | 1593 | L = numpy.where(L<0,1,L) |
|
|
1594 | 1594 | L = numpy.log(L) |
|
|
1595 | 1595 | tmp = numpy.sqrt(numpy.absolute(L)) |
|
|
1596 | 1596 | data_specwidth = (self.lambda_/(2*math.sqrt(2)*math.pi*self.ippSec*self.nCohInt))*tmp*numpy.sign(L) |
|
|
1597 | 1597 | n = self.__profIndex |
|
|
1598 | 1598 | |
|
|
1599 | 1599 | self.__buffer = numpy.zeros((self.__nch, self.__nProf,self.__nHeis), dtype='complex') |
|
|
1600 | 1600 | self.__profIndex = 0 |
|
|
1601 | 1601 | return data_power,data_intensity,data_velocity,data_snrPP,data_specwidth,n |
|
|
1602 | 1602 | |
|
|
1603 | 1603 | |
|
|
1604 | 1604 | def pulsePairbyProfiles(self,dataOut): |
|
|
1605 | 1605 | |
|
|
1606 | 1606 | self.__dataReady = False |
|
|
1607 | 1607 | data_power = None |
|
|
1608 | 1608 | data_intensity = None |
|
|
1609 | 1609 | data_velocity = None |
|
|
1610 | 1610 | data_specwidth = None |
|
|
1611 | 1611 | data_snrPP = None |
|
|
1612 | 1612 | self.putData(data=dataOut.data) |
|
|
1613 | 1613 | if self.__profIndex == self.n: |
|
|
1614 | 1614 | data_power,data_intensity, data_velocity,data_snrPP,data_specwidth, n = self.pushData(dataOut=dataOut) |
|
|
1615 | 1615 | self.__dataReady = True |
|
|
1616 | 1616 | |
|
|
1617 | 1617 | return data_power, data_intensity, data_velocity, data_snrPP, data_specwidth |
|
|
1618 | 1618 | |
|
|
1619 | 1619 | |
|
|
1620 | 1620 | def pulsePairOp(self, dataOut, datatime= None): |
|
|
1621 | 1621 | |
|
|
1622 | 1622 | if self.__initime == None: |
|
|
1623 | 1623 | self.__initime = datatime |
|
|
1624 | 1624 | data_power, data_intensity, data_velocity, data_snrPP, data_specwidth = self.pulsePairbyProfiles(dataOut) |
|
|
1625 | 1625 | self.__lastdatatime = datatime |
|
|
1626 | 1626 | |
|
|
1627 | 1627 | if data_power is None: |
|
|
1628 | 1628 | return None, None, None,None,None,None |
|
|
1629 | 1629 | |
|
|
1630 | 1630 | avgdatatime = self.__initime |
|
|
1631 | 1631 | deltatime = datatime - self.__lastdatatime |
|
|
1632 | 1632 | self.__initime = datatime |
|
|
1633 | 1633 | |
|
|
1634 | 1634 | return data_power, data_intensity, data_velocity, data_snrPP, data_specwidth, avgdatatime |
|
|
1635 | 1635 | |
|
|
1636 | 1636 | def run(self, dataOut,n = None,removeDC= False, overlapping= False,**kwargs): |
|
|
1637 | 1637 | |
|
|
1638 | 1638 | if not self.isConfig: |
|
|
1639 | 1639 | self.setup(dataOut = dataOut, n = n , removeDC=removeDC , **kwargs) |
|
|
1640 | 1640 | self.isConfig = True |
|
|
1641 | 1641 | data_power, data_intensity, data_velocity,data_snrPP,data_specwidth, avgdatatime = self.pulsePairOp(dataOut, dataOut.utctime) |
|
|
1642 | 1642 | dataOut.flagNoData = True |
|
|
1643 | 1643 | |
|
|
1644 | 1644 | if self.__dataReady: |
|
|
1645 | 1645 | dataOut.nCohInt *= self.n |
|
|
1646 | 1646 | dataOut.dataPP_POW = data_intensity # S |
|
|
1647 | 1647 | dataOut.dataPP_POWER = data_power # P |
|
|
1648 | 1648 | dataOut.dataPP_DOP = data_velocity |
|
|
1649 | 1649 | dataOut.dataPP_SNR = data_snrPP |
|
|
1650 | 1650 | dataOut.dataPP_WIDTH = data_specwidth |
|
|
1651 | 1651 | dataOut.PRFbyAngle = self.n #numero de PRF*cada angulo rotado que equivale a un tiempo. |
|
|
1652 | 1652 | dataOut.utctime = avgdatatime |
|
|
1653 | 1653 | dataOut.flagNoData = False |
|
|
1654 | 1654 | return dataOut |
|
|
1655 | 1655 | |
|
|
1656 | 1656 | |
|
|
1657 | 1657 | |
|
|
1658 | 1658 | # import collections |
|
|
1659 | 1659 | # from scipy.stats import mode |
|
|
1660 | 1660 | # |
|
|
1661 | 1661 | # class Synchronize(Operation): |
|
|
1662 | 1662 | # |
|
|
1663 | 1663 | # isConfig = False |
|
|
1664 | 1664 | # __profIndex = 0 |
|
|
1665 | 1665 | # |
|
|
1666 | 1666 | # def __init__(self, **kwargs): |
|
|
1667 | 1667 | # |
|
|
1668 | 1668 | # Operation.__init__(self, **kwargs) |
|
|
1669 | 1669 | # # self.isConfig = False |
|
|
1670 | 1670 | # self.__powBuffer = None |
|
|
1671 | 1671 | # self.__startIndex = 0 |
|
|
1672 | 1672 | # self.__pulseFound = False |
|
|
1673 | 1673 | # |
|
|
1674 | 1674 | # def __findTxPulse(self, dataOut, channel=0, pulse_with = None): |
|
|
1675 | 1675 | # |
|
|
1676 | 1676 | # #Read data |
|
|
1677 | 1677 | # |
|
|
1678 | 1678 | # powerdB = dataOut.getPower(channel = channel) |
|
|
1679 | 1679 | # noisedB = dataOut.getNoise(channel = channel)[0] |
|
|
1680 | 1680 | # |
|
|
1681 | 1681 | # self.__powBuffer.extend(powerdB.flatten()) |
|
|
1682 | 1682 | # |
|
|
1683 | 1683 | # dataArray = numpy.array(self.__powBuffer) |
|
|
1684 | 1684 | # |
|
|
1685 | 1685 | # filteredPower = numpy.correlate(dataArray, dataArray[0:self.__nSamples], "same") |
|
|
1686 | 1686 | # |
|
|
1687 | 1687 | # maxValue = numpy.nanmax(filteredPower) |
|
|
1688 | 1688 | # |
|
|
1689 | 1689 | # if maxValue < noisedB + 10: |
|
|
1690 | 1690 | # #No se encuentra ningun pulso de transmision |
|
|
1691 | 1691 | # return None |
|
|
1692 | 1692 | # |
|
|
1693 | 1693 | # maxValuesIndex = numpy.where(filteredPower > maxValue - 0.1*abs(maxValue))[0] |
|
|
1694 | 1694 | # |
|
|
1695 | 1695 | # if len(maxValuesIndex) < 2: |
|
|
1696 | 1696 | # #Solo se encontro un solo pulso de transmision de un baudio, esperando por el siguiente TX |
|
|
1697 | 1697 | # return None |
|
|
1698 | 1698 | # |
|
|
1699 | 1699 | # phasedMaxValuesIndex = maxValuesIndex - self.__nSamples |
|
|
1700 | 1700 | # |
|
|
1701 | 1701 | # #Seleccionar solo valores con un espaciamiento de nSamples |
|
|
1702 | 1702 | # pulseIndex = numpy.intersect1d(maxValuesIndex, phasedMaxValuesIndex) |
|
|
1703 | 1703 | # |
|
|
1704 | 1704 | # if len(pulseIndex) < 2: |
|
|
1705 | 1705 | # #Solo se encontro un pulso de transmision con ancho mayor a 1 |
|
|
1706 | 1706 | # return None |
|
|
1707 | 1707 | # |
|
|
1708 | 1708 | # spacing = pulseIndex[1:] - pulseIndex[:-1] |
|
|
1709 | 1709 | # |
|
|
1710 | 1710 | # #remover senales que se distancien menos de 10 unidades o muestras |
|
|
1711 | 1711 | # #(No deberian existir IPP menor a 10 unidades) |
|
|
1712 | 1712 | # |
|
|
1713 | 1713 | # realIndex = numpy.where(spacing > 10 )[0] |
|
|
1714 | 1714 | # |
|
|
1715 | 1715 | # if len(realIndex) < 2: |
|
|
1716 | 1716 | # #Solo se encontro un pulso de transmision con ancho mayor a 1 |
|
|
1717 | 1717 | # return None |
|
|
1718 | 1718 | # |
|
|
1719 | 1719 | # #Eliminar pulsos anchos (deja solo la diferencia entre IPPs) |
|
|
1720 | 1720 | # realPulseIndex = pulseIndex[realIndex] |
|
|
1721 | 1721 | # |
|
|
1722 | 1722 | # period = mode(realPulseIndex[1:] - realPulseIndex[:-1])[0][0] |
|
|
1723 | 1723 | # |
|
|
1724 | 1724 | # print "IPP = %d samples" %period |
|
|
1725 | 1725 | # |
|
|
1726 | 1726 | # self.__newNSamples = dataOut.nHeights #int(period) |
|
|
1727 | 1727 | # self.__startIndex = int(realPulseIndex[0]) |
|
|
1728 | 1728 | # |
|
|
1729 | 1729 | # return 1 |
|
|
1730 | 1730 | # |
|
|
1731 | 1731 | # |
|
|
1732 | 1732 | # def setup(self, nSamples, nChannels, buffer_size = 4): |
|
|
1733 | 1733 | # |
|
|
1734 | 1734 | # self.__powBuffer = collections.deque(numpy.zeros( buffer_size*nSamples,dtype=numpy.float), |
|
|
1735 | 1735 | # maxlen = buffer_size*nSamples) |
|
|
1736 | 1736 | # |
|
|
1737 | 1737 | # bufferList = [] |
|
|
1738 | 1738 | # |
|
|
1739 | 1739 | # for i in range(nChannels): |
|
|
1740 | 1740 | # bufferByChannel = collections.deque(numpy.zeros( buffer_size*nSamples, dtype=numpy.complex) + numpy.NAN, |
|
|
1741 | 1741 | # maxlen = buffer_size*nSamples) |
|
|
1742 | 1742 | # |
|
|
1743 | 1743 | # bufferList.append(bufferByChannel) |
|
|
1744 | 1744 | # |
|
|
1745 | 1745 | # self.__nSamples = nSamples |
|
|
1746 | 1746 | # self.__nChannels = nChannels |
|
|
1747 | 1747 | # self.__bufferList = bufferList |
|
|
1748 | 1748 | # |
|
|
1749 | 1749 | # def run(self, dataOut, channel = 0): |
|
|
1750 | 1750 | # |
|
|
1751 | 1751 | # if not self.isConfig: |
|
|
1752 | 1752 | # nSamples = dataOut.nHeights |
|
|
1753 | 1753 | # nChannels = dataOut.nChannels |
|
|
1754 | 1754 | # self.setup(nSamples, nChannels) |
|
|
1755 | 1755 | # self.isConfig = True |
|
|
1756 | 1756 | # |
|
|
1757 | 1757 | # #Append new data to internal buffer |
|
|
1758 | 1758 | # for thisChannel in range(self.__nChannels): |
|
|
1759 | 1759 | # bufferByChannel = self.__bufferList[thisChannel] |
|
|
1760 | 1760 | # bufferByChannel.extend(dataOut.data[thisChannel]) |
|
|
1761 | 1761 | # |
|
|
1762 | 1762 | # if self.__pulseFound: |
|
|
1763 | 1763 | # self.__startIndex -= self.__nSamples |
|
|
1764 | 1764 | # |
|
|
1765 | 1765 | # #Finding Tx Pulse |
|
|
1766 | 1766 | # if not self.__pulseFound: |
|
|
1767 | 1767 | # indexFound = self.__findTxPulse(dataOut, channel) |
|
|
1768 | 1768 | # |
|
|
1769 | 1769 | # if indexFound == None: |
|
|
1770 | 1770 | # dataOut.flagNoData = True |
|
|
1771 | 1771 | # return |
|
|
1772 | 1772 | # |
|
|
1773 | 1773 | # self.__arrayBuffer = numpy.zeros((self.__nChannels, self.__newNSamples), dtype = numpy.complex) |
|
|
1774 | 1774 | # self.__pulseFound = True |
|
|
1775 | 1775 | # self.__startIndex = indexFound |
|
|
1776 | 1776 | # |
|
|
1777 | 1777 | # #If pulse was found ... |
|
|
1778 | 1778 | # for thisChannel in range(self.__nChannels): |
|
|
1779 | 1779 | # bufferByChannel = self.__bufferList[thisChannel] |
|
|
1780 | 1780 | # #print self.__startIndex |
|
|
1781 | 1781 | # x = numpy.array(bufferByChannel) |
|
|
1782 | 1782 | # self.__arrayBuffer[thisChannel] = x[self.__startIndex:self.__startIndex+self.__newNSamples] |
|
|
1783 | 1783 | # |
|
|
1784 | 1784 | # deltaHeight = dataOut.heightList[1] - dataOut.heightList[0] |
|
|
1785 | 1785 | # dataOut.heightList = numpy.arange(self.__newNSamples)*deltaHeight |
|
|
1786 | 1786 | # # dataOut.ippSeconds = (self.__newNSamples / deltaHeight)/1e6 |
|
|
1787 | 1787 | # |
|
|
1788 | 1788 | # dataOut.data = self.__arrayBuffer |
|
|
1789 | 1789 | # |
|
|
1790 | 1790 | # self.__startIndex += self.__newNSamples |
|
|
1791 | 1791 | # |
|
|
1792 | 1792 | # return |
|
|
1793 | 1793 | class SSheightProfiles(Operation): |
|
|
1794 | 1794 | |
|
|
1795 | 1795 | step = None |
|
|
1796 | 1796 | nsamples = None |
|
|
1797 | 1797 | bufferShape = None |
|
|
1798 | 1798 | profileShape = None |
|
|
1799 | 1799 | sshProfiles = None |
|
|
1800 | 1800 | profileIndex = None |
|
|
1801 | 1801 | |
|
|
1802 | 1802 | def __init__(self, **kwargs): |
|
|
1803 | 1803 | |
|
|
1804 | 1804 | Operation.__init__(self, **kwargs) |
|
|
1805 | 1805 | self.isConfig = False |
|
|
1806 | 1806 | |
|
|
1807 | 1807 | def setup(self,dataOut ,step = None , nsamples = None): |
|
|
1808 | 1808 | |
|
|
1809 | 1809 | if step == None and nsamples == None: |
|
|
1810 | 1810 | raise ValueError("step or nheights should be specified ...") |
|
|
1811 | 1811 | |
|
|
1812 | 1812 | self.step = step |
|
|
1813 | 1813 | self.nsamples = nsamples |
|
|
1814 | 1814 | self.__nChannels = dataOut.nChannels |
|
|
1815 | 1815 | self.__nProfiles = dataOut.nProfiles |
|
|
1816 | 1816 | self.__nHeis = dataOut.nHeights |
|
|
1817 | 1817 | shape = dataOut.data.shape #nchannels, nprofiles, nsamples |
|
|
1818 | 1818 | |
|
|
1819 | 1819 | residue = (shape[1] - self.nsamples) % self.step |
|
|
1820 | 1820 | if residue != 0: |
|
|
1821 | 1821 | print("The residue is %d, step=%d should be multiple of %d to avoid loss of %d samples"%(residue,step,shape[1] - self.nsamples,residue)) |
|
|
1822 | 1822 | |
|
|
1823 | 1823 | deltaHeight = dataOut.heightList[1] - dataOut.heightList[0] |
|
|
1824 | 1824 | numberProfile = self.nsamples |
|
|
1825 | 1825 | numberSamples = (shape[1] - self.nsamples)/self.step |
|
|
1826 | 1826 | |
|
|
1827 | 1827 | self.bufferShape = int(shape[0]), int(numberSamples), int(numberProfile) # nchannels, nsamples , nprofiles |
|
|
1828 | 1828 | self.profileShape = int(shape[0]), int(numberProfile), int(numberSamples) # nchannels, nprofiles, nsamples |
|
|
1829 | 1829 | |
|
|
1830 | 1830 | self.buffer = numpy.zeros(self.bufferShape , dtype=numpy.complex) |
|
|
1831 | 1831 | self.sshProfiles = numpy.zeros(self.profileShape, dtype=numpy.complex) |
|
|
1832 | 1832 | |
|
|
1833 | 1833 | def run(self, dataOut, step, nsamples, code = None, repeat = None): |
|
|
1834 | 1834 | dataOut.flagNoData = True |
|
|
1835 | 1835 | |
|
|
1836 | 1836 | profileIndex = None |
|
|
1837 | 1837 | #print("nProfiles, nHeights ",dataOut.nProfiles, dataOut.nHeights) |
|
|
1838 | 1838 | #print(dataOut.getFreqRange(1)/1000.) |
|
|
1839 | 1839 | #exit(1) |
|
|
1840 | 1840 | if dataOut.flagDataAsBlock: |
|
|
1841 | 1841 | dataOut.data = numpy.average(dataOut.data,axis=1) |
|
|
1842 | 1842 | #print("jee") |
|
|
1843 | 1843 | dataOut.flagDataAsBlock = False |
|
|
1844 | 1844 | if not self.isConfig: |
|
|
1845 | 1845 | self.setup(dataOut, step=step , nsamples=nsamples) |
|
|
1846 | 1846 | #print("Setup done") |
|
|
1847 | 1847 | self.isConfig = True |
|
|
1848 | 1848 | |
|
|
1849 | 1849 | |
|
|
1850 | 1850 | if code is not None: |
|
|
1851 | 1851 | code = numpy.array(code) |
|
|
1852 | 1852 | code_block = code |
|
|
1853 | 1853 | |
|
|
1854 | 1854 | if repeat is not None: |
|
|
1855 | 1855 | code_block = numpy.repeat(code_block, repeats=repeat, axis=1) |
|
|
1856 | 1856 | #print(code_block.shape) |
|
|
1857 | 1857 | for i in range(self.buffer.shape[1]): |
|
|
1858 | 1858 | |
|
|
1859 | 1859 | if code is not None: |
|
|
1860 | 1860 | self.buffer[:,i] = dataOut.data[:,i*self.step:i*self.step + self.nsamples]*code_block |
|
|
1861 | 1861 | |
|
|
1862 | 1862 | else: |
|
|
1863 | 1863 | |
|
|
1864 | 1864 | self.buffer[:,i] = dataOut.data[:,i*self.step:i*self.step + self.nsamples]#*code[dataOut.profileIndex,:] |
|
|
1865 | 1865 | |
|
|
1866 | 1866 | #self.buffer[:,j,self.__nHeis-j*self.step - self.nheights:self.__nHeis-j*self.step] = numpy.flip(dataOut.data[:,j*self.step:j*self.step + self.nheights]) |
|
|
1867 | 1867 | |
|
|
1868 | 1868 | for j in range(self.buffer.shape[0]): |
|
|
1869 | 1869 | self.sshProfiles[j] = numpy.transpose(self.buffer[j]) |
|
|
1870 | 1870 | |
|
|
1871 | 1871 | profileIndex = self.nsamples |
|
|
1872 | 1872 | deltaHeight = dataOut.heightList[1] - dataOut.heightList[0] |
|
|
1873 | 1873 | ippSeconds = (deltaHeight*1.0e-6)/(0.15) |
|
|
1874 | 1874 | #print("ippSeconds, dH: ",ippSeconds,deltaHeight) |
|
|
1875 | 1875 | try: |
|
|
1876 | 1876 | if dataOut.concat_m is not None: |
|
|
1877 | 1877 | ippSeconds= ippSeconds/float(dataOut.concat_m) |
|
|
1878 | 1878 | #print "Profile concat %d"%dataOut.concat_m |
|
|
1879 | 1879 | except: |
|
|
1880 | 1880 | pass |
|
|
1881 | 1881 | |
|
|
1882 | 1882 | dataOut.data = self.sshProfiles |
|
|
1883 | 1883 | dataOut.flagNoData = False |
|
|
1884 | 1884 | dataOut.heightList = numpy.arange(self.buffer.shape[1]) *self.step*deltaHeight + dataOut.heightList[0] |
|
|
1885 | 1885 | dataOut.nProfiles = int(dataOut.nProfiles*self.nsamples) |
|
|
1886 | 1886 | |
|
|
1887 | 1887 | dataOut.profileIndex = profileIndex |
|
|
1888 | 1888 | dataOut.flagDataAsBlock = True |
|
|
1889 | 1889 | dataOut.ippSeconds = ippSeconds |
|
|
1890 | 1890 | dataOut.step = self.step |
|
|
1891 | 1891 | #print(numpy.shape(dataOut.data)) |
|
|
1892 | 1892 | #exit(1) |
|
|
1893 | 1893 | #print("new data shape and time:", dataOut.data.shape, dataOut.utctime) |
|
|
1894 | 1894 | |
|
|
1895 | 1895 | return dataOut |
|
|
1896 | 1896 | ################################################################################3############################3 |
|
|
1897 | 1897 | ################################################################################3############################3 |
|
|
1898 | 1898 | ################################################################################3############################3 |
|
|
1899 | 1899 | ################################################################################3############################3 |
|
|
1900 | 1900 | |
|
|
1901 | 1901 | class SSheightProfiles2(Operation): |
|
|
1902 | 1902 | ''' |
|
|
1903 | 1903 | Procesa por perfiles y por bloques |
|
|
1904 | 1904 | VersiΓ³n corregida y actualizada para trabajar con RemoveProfileSats2 |
|
|
1905 | 1905 | Usar esto |
|
|
1906 | 1906 | ''' |
|
|
1907 | 1907 | |
|
|
1908 | 1908 | |
|
|
1909 | 1909 | bufferShape = None |
|
|
1910 | 1910 | profileShape = None |
|
|
1911 | 1911 | sshProfiles = None |
|
|
1912 | 1912 | profileIndex = None |
|
|
1913 | 1913 | #nsamples = None |
|
|
1914 | 1914 | #step = None |
|
|
1915 | 1915 | #deltaHeight = None |
|
|
1916 | 1916 | #init_range = None |
|
|
1917 | 1917 | __slots__ = ('step', 'nsamples', 'deltaHeight', 'init_range', 'isConfig', '__nChannels', |
|
|
1918 | 1918 | '__nProfiles', '__nHeis', 'deltaHeight', 'new_nHeights') |
|
|
1919 | 1919 | |
|
|
1920 | 1920 | def __init__(self, **kwargs): |
|
|
1921 | 1921 | |
|
|
1922 | 1922 | Operation.__init__(self, **kwargs) |
|
|
1923 | 1923 | self.isConfig = False |
|
|
1924 | 1924 | |
|
|
1925 | 1925 | def setup(self,dataOut ,step = None , nsamples = None): |
|
|
1926 | 1926 | |
|
|
1927 | 1927 | if step == None and nsamples == None: |
|
|
1928 | 1928 | raise ValueError("step or nheights should be specified ...") |
|
|
1929 | 1929 | |
|
|
1930 | 1930 | self.step = step |
|
|
1931 | 1931 | self.nsamples = nsamples |
|
|
1932 | 1932 | self.__nChannels = int(dataOut.nChannels) |
|
|
1933 | 1933 | self.__nProfiles = int(dataOut.nProfiles) |
|
|
1934 | 1934 | self.__nHeis = int(dataOut.nHeights) |
|
|
1935 | 1935 | |
|
|
1936 | 1936 | residue = (self.__nHeis - self.nsamples) % self.step |
|
|
1937 | 1937 | if residue != 0: |
|
|
1938 | 1938 | print("The residue is %d, step=%d should be multiple of %d to avoid loss of %d samples"%(residue,step,self.__nProfiles - self.nsamples,residue)) |
|
|
1939 | 1939 | |
|
|
1940 | 1940 | self.deltaHeight = dataOut.heightList[1] - dataOut.heightList[0] |
|
|
1941 | 1941 | self.init_range = dataOut.heightList[0] |
|
|
1942 | 1942 | #numberProfile = self.nsamples |
|
|
1943 | 1943 | numberSamples = (self.__nHeis - self.nsamples)/self.step |
|
|
1944 | 1944 | |
|
|
1945 | 1945 | self.new_nHeights = numberSamples |
|
|
1946 | 1946 | |
|
|
1947 | 1947 | self.bufferShape = int(self.__nChannels), int(numberSamples), int(self.nsamples) # nchannels, nsamples , nprofiles |
|
|
1948 | 1948 | self.profileShape = int(self.__nChannels), int(self.nsamples), int(numberSamples) # nchannels, nprofiles, nsamples |
|
|
1949 | 1949 | |
|
|
1950 | 1950 | self.buffer = numpy.zeros(self.bufferShape , dtype=numpy.complex) |
|
|
1951 | 1951 | self.sshProfiles = numpy.zeros(self.profileShape, dtype=numpy.complex) |
|
|
1952 | 1952 | |
|
|
1953 | 1953 | def getNewProfiles(self, data, code=None, repeat=None): |
|
|
1954 | 1954 | |
|
|
1955 | 1955 | if code is not None: |
|
|
1956 | 1956 | code = numpy.array(code) |
|
|
1957 | 1957 | code_block = code |
|
|
1958 | 1958 | |
|
|
1959 | 1959 | if repeat is not None: |
|
|
1960 | 1960 | code_block = numpy.repeat(code_block, repeats=repeat, axis=1) |
|
|
1961 | 1961 | if data.ndim < 3: |
|
|
1962 | 1962 | data = data.reshape(self.__nChannels,1,self.__nHeis ) |
|
|
1963 | 1963 | #print("buff, data, :",self.buffer.shape, data.shape,self.sshProfiles.shape, code_block.shape) |
|
|
1964 | 1964 | for ch in range(self.__nChannels): |
|
|
1965 | 1965 | for i in range(int(self.new_nHeights)): #nuevas alturas |
|
|
1966 | 1966 | if code is not None: |
|
|
1967 | 1967 | self.buffer[ch,i,:] = data[ch,:,i*self.step:i*self.step + self.nsamples]*code_block |
|
|
1968 | 1968 | else: |
|
|
1969 | 1969 | self.buffer[ch,i,:] = data[ch,:,i*self.step:i*self.step + self.nsamples]#*code[dataOut.profileIndex,:] |
|
|
1970 | 1970 | |
|
|
1971 | 1971 | for j in range(self.__nChannels): #en los cananles |
|
|
1972 | 1972 | self.sshProfiles[j,:,:] = numpy.transpose(self.buffer[j,:,:]) |
|
|
1973 | 1973 | #print("new profs Done") |
|
|
1974 | 1974 | |
|
|
1975 | 1975 | |
|
|
1976 | 1976 | |
|
|
1977 | 1977 | def run(self, dataOut, step, nsamples, code = None, repeat = None): |
|
|
1978 | 1978 | # print("running") |
|
|
1979 | 1979 | if dataOut.flagNoData == True: |
|
|
1980 | 1980 | return dataOut |
|
|
1981 | 1981 | dataOut.flagNoData = True |
|
|
1982 | 1982 | #print("init data shape:", dataOut.data.shape) |
|
|
1983 | 1983 | #print("ch: {} prof: {} hs: {}".format(int(dataOut.nChannels), |
|
|
1984 | 1984 | # int(dataOut.nProfiles),int(dataOut.nHeights))) |
|
|
1985 | 1985 | |
|
|
1986 | 1986 | profileIndex = None |
|
|
1987 | 1987 | # if not dataOut.flagDataAsBlock: |
|
|
1988 | 1988 | # dataOut.nProfiles = 1 |
|
|
1989 | 1989 | |
|
|
1990 | 1990 | if not self.isConfig: |
|
|
1991 | 1991 | self.setup(dataOut, step=step , nsamples=nsamples) |
|
|
1992 | 1992 | #print("Setup done") |
|
|
1993 | 1993 | self.isConfig = True |
|
|
1994 | 1994 | |
|
|
1995 | 1995 | dataBlock = None |
|
|
1996 | 1996 | |
|
|
1997 | 1997 | nprof = 1 |
|
|
1998 | 1998 | if dataOut.flagDataAsBlock: |
|
|
1999 | 1999 | nprof = int(dataOut.nProfiles) |
|
|
2000 | 2000 | |
|
|
2001 | 2001 | #print("dataOut nProfiles:", dataOut.nProfiles) |
|
|
2002 | 2002 | for profile in range(nprof): |
|
|
2003 | 2003 | if dataOut.flagDataAsBlock: |
|
|
2004 | 2004 | #print("read blocks") |
|
|
2005 | 2005 | self.getNewProfiles(dataOut.data[:,profile,:], code=code, repeat=repeat) |
|
|
2006 | 2006 | else: |
|
|
2007 | 2007 | #print("read profiles") |
|
|
2008 | 2008 | self.getNewProfiles(dataOut.data, code=code, repeat=repeat) #only one channe |
|
|
2009 | 2009 | if profile == 0: |
|
|
2010 | 2010 | dataBlock = self.sshProfiles.copy() |
|
|
2011 | 2011 | else: #by blocks |
|
|
2012 | 2012 | dataBlock = numpy.concatenate((dataBlock,self.sshProfiles), axis=1) #profile axis |
|
|
2013 | 2013 | #print("by blocks: ",dataBlock.shape, self.sshProfiles.shape) |
|
|
2014 | 2014 | |
|
|
2015 | 2015 | profileIndex = self.nsamples |
|
|
2016 | 2016 | #deltaHeight = dataOut.heightList[1] - dataOut.heightList[0] |
|
|
2017 | 2017 | ippSeconds = (self.deltaHeight*1.0e-6)/(0.15) |
|
|
2018 | 2018 | |
|
|
2019 | 2019 | |
|
|
2020 | 2020 | dataOut.data = dataBlock |
|
|
2021 | 2021 | #print("show me: ",self.step,self.deltaHeight, dataOut.heightList, self.new_nHeights) |
|
|
2022 | 2022 | dataOut.heightList = numpy.arange(int(self.new_nHeights)) *self.step*self.deltaHeight + self.init_range |
|
|
2023 | 2023 | dataOut.sampled_heightsFFT = self.nsamples |
|
|
2024 | 2024 | dataOut.ippSeconds = ippSeconds |
|
|
2025 | 2025 | dataOut.step = self.step |
|
|
2026 | 2026 | dataOut.deltaHeight = self.step*self.deltaHeight |
|
|
2027 | 2027 | dataOut.flagNoData = False |
|
|
2028 | 2028 | if dataOut.flagDataAsBlock: |
|
|
2029 | 2029 | dataOut.nProfiles = int(dataOut.nProfiles*self.nsamples) |
|
|
2030 | 2030 | |
|
|
2031 | 2031 | else: |
|
|
2032 | 2032 | dataOut.nProfiles = int(self.nsamples) |
|
|
2033 | 2033 | dataOut.profileIndex = dataOut.nProfiles |
|
|
2034 | 2034 | dataOut.flagDataAsBlock = True |
|
|
2035 | 2035 | |
|
|
2036 | 2036 | dataBlock = None |
|
|
2037 | 2037 | |
|
|
2038 | 2038 | #print("new data shape:", dataOut.data.shape, dataOut.utctime) |
|
|
2039 | 2039 | |
|
|
2040 | 2040 | #update Processing Header: |
|
|
2041 | 2041 | dataOut.processingHeaderObj.heightList = dataOut.heightList |
|
|
2042 | 2042 | dataOut.processingHeaderObj.ipp = ippSeconds |
|
|
2043 | 2043 | dataOut.processingHeaderObj.heightResolution = dataOut.deltaHeight |
|
|
2044 | 2044 | #dataOut.processingHeaderObj.profilesPerBlock = nProfiles |
|
|
2045 | 2045 | |
|
|
2046 | 2046 | # # dataOut.data = CH, PROFILES, HEIGHTS |
|
|
2047 | 2047 | #print(dataOut.data .shape) |
|
|
2048 | 2048 | if dataOut.flagProfilesByRange: |
|
|
2049 | 2049 | # #assuming the same remotion for all channels |
|
|
2050 | 2050 | aux = [ self.nsamples - numpy.count_nonzero(dataOut.data[0, :, h]==0) for h in range(len(dataOut.heightList))] |
|
|
2051 | 2051 | dataOut.nProfilesByRange = (numpy.asarray(aux)).reshape((1,len(dataOut.heightList) )) |
|
|
2052 | 2052 | #print(dataOut.nProfilesByRange.shape) |
|
|
2053 | 2053 | else: |
|
|
2054 | 2054 | dataOut.nProfilesByRange = numpy.ones((1, len(dataOut.heightList)))*dataOut.nProfiles |
|
|
2055 | 2055 | return dataOut |
|
|
2056 | 2056 | |
|
|
2057 | 2057 | |
|
|
2058 | 2058 | |
|
|
2059 | 2059 | class RemoveProfileSats(Operation): |
|
|
2060 | 2060 | ''' |
|
|
2061 | 2061 | Escrito: Joab Apaza |
|
|
2062 | 2062 | |
|
|
2063 | 2063 | Omite los perfiles contaminados con seΓ±al de satΓ©lites, usando una altura de referencia |
|
|
2064 | 2064 | In: minHei = min_sat_range |
|
|
2065 | 2065 | max_sat_range |
|
|
2066 | 2066 | min_hei_ref |
|
|
2067 | 2067 | max_hei_ref |
|
|
2068 | 2068 | th = diference between profiles mean, ref and sats |
|
|
2069 | 2069 | Out: |
|
|
2070 | 2070 | profile clean |
|
|
2071 | 2071 | ''' |
|
|
2072 | 2072 | |
|
|
2073 | 2073 | |
|
|
2074 | 2074 | __buffer_data = [] |
|
|
2075 | 2075 | __buffer_times = [] |
|
|
2076 | 2076 | |
|
|
2077 | 2077 | buffer = None |
|
|
2078 | 2078 | |
|
|
2079 | 2079 | outliers_IDs_list = [] |
|
|
2080 | 2080 | |
|
|
2081 | 2081 | |
|
|
2082 | 2082 | __slots__ = ('n','navg','profileMargin','thHistOutlier','minHei_idx','maxHei_idx','nHeights', |
|
|
2083 | 2083 | 'first_utcBlock','__profIndex','init_prof','end_prof','lenProfileOut','nChannels', |
|
|
2084 | 2084 | '__count_exec','__initime','__dataReady','__ipp', 'minRef', 'maxRef', 'thdB') |
|
|
2085 | 2085 | def __init__(self, **kwargs): |
|
|
2086 | 2086 | |
|
|
2087 | 2087 | Operation.__init__(self, **kwargs) |
|
|
2088 | 2088 | self.isConfig = False |
|
|
2089 | 2089 | |
|
|
2090 | 2090 | def setup(self,dataOut, n=None , navg=0.8, profileMargin=50,thHistOutlier=15, |
|
|
2091 | 2091 | minHei=None, maxHei=None, minRef=None, maxRef=None, thdB=10): |
|
|
2092 | 2092 | |
|
|
2093 | 2093 | if n == None and timeInterval == None: |
|
|
2094 | 2094 | raise ValueError("nprofiles or timeInterval should be specified ...") |
|
|
2095 | 2095 | |
|
|
2096 | 2096 | if n != None: |
|
|
2097 | 2097 | self.n = n |
|
|
2098 | 2098 | |
|
|
2099 | 2099 | self.navg = navg |
|
|
2100 | 2100 | self.profileMargin = profileMargin |
|
|
2101 | 2101 | self.thHistOutlier = thHistOutlier |
|
|
2102 | 2102 | self.__profIndex = 0 |
|
|
2103 | 2103 | self.buffer = None |
|
|
2104 | 2104 | self._ipp = dataOut.ippSeconds |
|
|
2105 | 2105 | self.n_prof_released = 0 |
|
|
2106 | 2106 | self.heightList = dataOut.heightList |
|
|
2107 | 2107 | self.init_prof = 0 |
|
|
2108 | 2108 | self.end_prof = 0 |
|
|
2109 | 2109 | self.__count_exec = 0 |
|
|
2110 | 2110 | self.__profIndex = 0 |
|
|
2111 | 2111 | self.first_utcBlock = None |
|
|
2112 | 2112 | #self.__dh = dataOut.heightList[1] - dataOut.heightList[0] |
|
|
2113 | 2113 | minHei = minHei |
|
|
2114 | 2114 | maxHei = maxHei |
|
|
2115 | 2115 | if minHei==None : |
|
|
2116 | 2116 | minHei = dataOut.heightList[0] |
|
|
2117 | 2117 | if maxHei==None : |
|
|
2118 | 2118 | maxHei = dataOut.heightList[-1] |
|
|
2119 | 2119 | self.minHei_idx,self.maxHei_idx = getHei_index(minHei, maxHei, dataOut.heightList) |
|
|
2120 | 2120 | self.min_ref, self.max_ref = getHei_index(minRef, maxRef, dataOut.heightList) |
|
|
2121 | 2121 | self.nChannels = dataOut.nChannels |
|
|
2122 | 2122 | self.nHeights = dataOut.nHeights |
|
|
2123 | 2123 | self.test_counter = 0 |
|
|
2124 | 2124 | self.thdB = thdB |
|
|
2125 | 2125 | |
|
|
2126 | 2126 | def filterSatsProfiles(self): |
|
|
2127 | 2127 | data = self.__buffer_data |
|
|
2128 | 2128 | #print(data.shape) |
|
|
2129 | 2129 | nChannels, profiles, heights = data.shape |
|
|
2130 | 2130 | indexes=numpy.zeros([], dtype=int) |
|
|
2131 | 2131 | outliers_IDs=[] |
|
|
2132 | 2132 | for c in range(nChannels): |
|
|
2133 | 2133 | #print(self.min_ref,self.max_ref) |
|
|
2134 | 2134 | noise_ref = 10* numpy.log10((data[c,:,self.min_ref:self.max_ref] * numpy.conjugate(data[c,:,self.min_ref:self.max_ref])).real) |
|
|
2135 | 2135 | #print("Noise ",numpy.percentile(noise_ref,95)) |
|
|
2136 | 2136 | p95 = numpy.percentile(noise_ref,95) |
|
|
2137 | 2137 | noise_ref = noise_ref.mean() |
|
|
2138 | 2138 | #print("Noise ",noise_ref |
|
|
2139 | 2139 | |
|
|
2140 | 2140 | |
|
|
2141 | 2141 | for h in range(self.minHei_idx, self.maxHei_idx): |
|
|
2142 | 2142 | power = 10* numpy.log10((data[c,:,h] * numpy.conjugate(data[c,:,h])).real) |
|
|
2143 | 2143 | #th = noise_ref + self.thdB |
|
|
2144 | 2144 | th = noise_ref + 1.5*(p95-noise_ref) |
|
|
2145 | 2145 | index = numpy.where(power > th ) |
|
|
2146 | 2146 | if index[0].size > 10 and index[0].size < int(self.navg*profiles): |
|
|
2147 | 2147 | indexes = numpy.append(indexes, index[0]) |
|
|
2148 | 2148 | #print(index[0]) |
|
|
2149 | 2149 | #print(index[0]) |
|
|
2150 | 2150 | |
|
|
2151 | 2151 | # fig,ax = plt.subplots() |
|
|
2152 | 2152 | # #ax.set_title(str(k)+" "+str(j)) |
|
|
2153 | 2153 | # x=range(len(power)) |
|
|
2154 | 2154 | # ax.scatter(x,power) |
|
|
2155 | 2155 | # #ax.axvline(index) |
|
|
2156 | 2156 | # plt.grid() |
|
|
2157 | 2157 | # plt.show() |
|
|
2158 | 2158 | #print(indexes) |
|
|
2159 | 2159 | |
|
|
2160 | 2160 | #outliers_IDs = outliers_IDs.astype(numpy.dtype('int64')) |
|
|
2161 | 2161 | #outliers_IDs = numpy.unique(outliers_IDs) |
|
|
2162 | 2162 | |
|
|
2163 | 2163 | outs_lines = numpy.unique(indexes) |
|
|
2164 | 2164 | |
|
|
2165 | 2165 | |
|
|
2166 | 2166 | #Agrupando el histograma de outliers, |
|
|
2167 | 2167 | my_bins = numpy.linspace(0,int(profiles), int(profiles/100), endpoint=True) |
|
|
2168 | 2168 | |
|
|
2169 | 2169 | |
|
|
2170 | 2170 | hist, bins = numpy.histogram(outs_lines,bins=my_bins) |
|
|
2171 | 2171 | hist_outliers_indexes = numpy.where(hist > self.thHistOutlier) #es outlier |
|
|
2172 | 2172 | hist_outliers_indexes = hist_outliers_indexes[0] |
|
|
2173 | 2173 | # if len(hist_outliers_indexes>0): |
|
|
2174 | 2174 | # hist_outliers_indexes = numpy.append(hist_outliers_indexes,hist_outliers_indexes[-1]+1) |
|
|
2175 | 2175 | #print(hist_outliers_indexes) |
|
|
2176 | 2176 | #print(bins, hist_outliers_indexes) |
|
|
2177 | 2177 | bins_outliers_indexes = [int(i) for i in (bins[hist_outliers_indexes])] # |
|
|
2178 | 2178 | outlier_loc_index = [] |
|
|
2179 | 2179 | # for n in range(len(bins_outliers_indexes)): |
|
|
2180 | 2180 | # for e in range(bins_outliers_indexes[n]-self.profileMargin,bins_outliers_indexes[n]+ self.profileMargin): |
|
|
2181 | 2181 | # outlier_loc_index.append(e) |
|
|
2182 | 2182 | outlier_loc_index = [e for n in range(len(bins_outliers_indexes)) for e in range(bins_outliers_indexes[n]-self.profileMargin,bins_outliers_indexes[n]+ profiles//100 + self.profileMargin) ] |
|
|
2183 | 2183 | outlier_loc_index = numpy.asarray(outlier_loc_index) |
|
|
2184 | 2184 | |
|
|
2185 | 2185 | |
|
|
2186 | 2186 | |
|
|
2187 | 2187 | |
|
|
2188 | 2188 | #print("outliers Ids: ", outlier_loc_index, outlier_loc_index.shape) |
|
|
2189 | 2189 | outlier_loc_index = outlier_loc_index[ (outlier_loc_index >= 0) & (outlier_loc_index<profiles)] |
|
|
2190 | 2190 | #print("outliers final: ", outlier_loc_index) |
|
|
2191 | 2191 | |
|
|
2192 | 2192 | from matplotlib import pyplot as plt |
|
|
2193 | 2193 | x, y = numpy.meshgrid(numpy.arange(profiles), self.heightList) |
|
|
2194 | 2194 | fig, ax = plt.subplots(1,2,figsize=(8, 6)) |
|
|
2195 | 2195 | dat = data[0,:,:].real |
|
|
2196 | 2196 | dat = 10* numpy.log10((data[0,:,:] * numpy.conjugate(data[0,:,:])).real) |
|
|
2197 | 2197 | m = numpy.nanmean(dat) |
|
|
2198 | 2198 | o = numpy.nanstd(dat) |
|
|
2199 | 2199 | #print(m, o, x.shape, y.shape) |
|
|
2200 | 2200 | #c = ax[0].pcolormesh(x, y, dat.T, cmap ='YlGnBu', vmin = (m-2*o), vmax = (m+2*o)) |
|
|
2201 | 2201 | c = ax[0].pcolormesh(x, y, dat.T, cmap ='YlGnBu', vmin = 50, vmax = 75) |
|
|
2202 | 2202 | ax[0].vlines(outs_lines,200,600, linestyles='dashed', label = 'outs', color='w') |
|
|
2203 | 2203 | fig.colorbar(c) |
|
|
2204 | 2204 | ax[0].vlines(outlier_loc_index,650,750, linestyles='dashed', label = 'outs', color='r') |
|
|
2205 | 2205 | ax[1].hist(outs_lines,bins=my_bins) |
|
|
2206 | 2206 | plt.show() |
|
|
2207 | 2207 | |
|
|
2208 | 2208 | |
|
|
2209 | 2209 | self.outliers_IDs_list = outlier_loc_index |
|
|
2210 | 2210 | #print("outs list: ", self.outliers_IDs_list) |
|
|
2211 | 2211 | return data |
|
|
2212 | 2212 | |
|
|
2213 | 2213 | |
|
|
2214 | 2214 | |
|
|
2215 | 2215 | def fillBuffer(self, data, datatime): |
|
|
2216 | 2216 | |
|
|
2217 | 2217 | if self.__profIndex == 0: |
|
|
2218 | 2218 | self.__buffer_data = data.copy() |
|
|
2219 | 2219 | |
|
|
2220 | 2220 | else: |
|
|
2221 | 2221 | self.__buffer_data = numpy.concatenate((self.__buffer_data,data), axis=1)#en perfiles |
|
|
2222 | 2222 | self.__profIndex += 1 |
|
|
2223 | 2223 | self.__buffer_times.append(datatime) |
|
|
2224 | 2224 | |
|
|
2225 | 2225 | def getData(self, data, datatime=None): |
|
|
2226 | 2226 | |
|
|
2227 | 2227 | if self.__profIndex == 0: |
|
|
2228 | 2228 | self.__initime = datatime |
|
|
2229 | 2229 | |
|
|
2230 | 2230 | |
|
|
2231 | 2231 | self.__dataReady = False |
|
|
2232 | 2232 | |
|
|
2233 | 2233 | self.fillBuffer(data, datatime) |
|
|
2234 | 2234 | dataBlock = None |
|
|
2235 | 2235 | |
|
|
2236 | 2236 | if self.__profIndex == self.n: |
|
|
2237 | 2237 | #print("apnd : ",data) |
|
|
2238 | 2238 | dataBlock = self.filterSatsProfiles() |
|
|
2239 | 2239 | self.__dataReady = True |
|
|
2240 | 2240 | |
|
|
2241 | 2241 | return dataBlock |
|
|
2242 | 2242 | |
|
|
2243 | 2243 | if dataBlock is None: |
|
|
2244 | 2244 | return None, None |
|
|
2245 | 2245 | |
|
|
2246 | 2246 | |
|
|
2247 | 2247 | |
|
|
2248 | 2248 | return dataBlock |
|
|
2249 | 2249 | |
|
|
2250 | 2250 | def releaseBlock(self): |
|
|
2251 | 2251 | |
|
|
2252 | 2252 | if self.n % self.lenProfileOut != 0: |
|
|
2253 | 2253 | raise ValueError("lenProfileOut %d must be submultiple of nProfiles %d" %(self.lenProfileOut, self.n)) |
|
|
2254 | 2254 | return None |
|
|
2255 | 2255 | |
|
|
2256 | 2256 | data = self.buffer[:,self.init_prof:self.end_prof:,:] #ch, prof, alt |
|
|
2257 | 2257 | |
|
|
2258 | 2258 | self.init_prof = self.end_prof |
|
|
2259 | 2259 | self.end_prof += self.lenProfileOut |
|
|
2260 | 2260 | #print("data release shape: ",dataOut.data.shape, self.end_prof) |
|
|
2261 | 2261 | self.n_prof_released += 1 |
|
|
2262 | 2262 | |
|
|
2263 | 2263 | return data |
|
|
2264 | 2264 | |
|
|
2265 | 2265 | def run(self, dataOut, n=None, navg=0.8, nProfilesOut=1, profile_margin=50, |
|
|
2266 | 2266 | th_hist_outlier=15,minHei=None, maxHei=None, minRef=None, maxRef=None, thdB=10): |
|
|
2267 | 2267 | |
|
|
2268 | 2268 | if not self.isConfig: |
|
|
2269 | 2269 | #print("init p idx: ", dataOut.profileIndex ) |
|
|
2270 | 2270 | self.setup(dataOut,n=n, navg=navg,profileMargin=profile_margin,thHistOutlier=th_hist_outlier, |
|
|
2271 | 2271 | minHei=minHei, maxHei=maxHei, minRef=minRef, maxRef=maxRef, thdB=thdB) |
|
|
2272 | 2272 | self.isConfig = True |
|
|
2273 | 2273 | |
|
|
2274 | 2274 | dataBlock = None |
|
|
2275 | 2275 | |
|
|
2276 | 2276 | if not dataOut.buffer_empty: #hay datos acumulados |
|
|
2277 | 2277 | |
|
|
2278 | 2278 | if self.init_prof == 0: |
|
|
2279 | 2279 | self.n_prof_released = 0 |
|
|
2280 | 2280 | self.lenProfileOut = nProfilesOut |
|
|
2281 | 2281 | dataOut.flagNoData = False |
|
|
2282 | 2282 | #print("tp 2 ",dataOut.data.shape) |
|
|
2283 | 2283 | |
|
|
2284 | 2284 | self.init_prof = 0 |
|
|
2285 | 2285 | self.end_prof = self.lenProfileOut |
|
|
2286 | 2286 | |
|
|
2287 | 2287 | dataOut.nProfiles = self.lenProfileOut |
|
|
2288 | 2288 | if nProfilesOut == 1: |
|
|
2289 | 2289 | dataOut.flagDataAsBlock = False |
|
|
2290 | 2290 | else: |
|
|
2291 | 2291 | dataOut.flagDataAsBlock = True |
|
|
2292 | 2292 | #print("prof: ",self.init_prof) |
|
|
2293 | 2293 | dataOut.flagNoData = False |
|
|
2294 | 2294 | if numpy.isin(self.n_prof_released, self.outliers_IDs_list): |
|
|
2295 | 2295 | #print("omitting: ", self.n_prof_released) |
|
|
2296 | 2296 | dataOut.flagNoData = True |
|
|
2297 | 2297 | dataOut.ippSeconds = self._ipp |
|
|
2298 | 2298 | dataOut.utctime = self.first_utcBlock + self.init_prof*self._ipp |
|
|
2299 | 2299 | # print("time: ", dataOut.utctime, self.first_utcBlock, self.init_prof,self._ipp,dataOut.ippSeconds) |
|
|
2300 | 2300 | #dataOut.data = self.releaseBlock() |
|
|
2301 | 2301 | #########################################################3 |
|
|
2302 | 2302 | if self.n % self.lenProfileOut != 0: |
|
|
2303 | 2303 | raise ValueError("lenProfileOut %d must be submultiple of nProfiles %d" %(self.lenProfileOut, self.n)) |
|
|
2304 | 2304 | return None |
|
|
2305 | 2305 | |
|
|
2306 | 2306 | dataOut.data = None |
|
|
2307 | 2307 | |
|
|
2308 | 2308 | if nProfilesOut == 1: |
|
|
2309 | 2309 | dataOut.data = self.buffer[:,self.end_prof-1,:] #ch, prof, alt |
|
|
2310 | 2310 | else: |
|
|
2311 | 2311 | dataOut.data = self.buffer[:,self.init_prof:self.end_prof,:] #ch, prof, alt |
|
|
2312 | 2312 | |
|
|
2313 | 2313 | self.init_prof = self.end_prof |
|
|
2314 | 2314 | self.end_prof += self.lenProfileOut |
|
|
2315 | 2315 | #print("data release shape: ",dataOut.data.shape, self.end_prof, dataOut.flagNoData) |
|
|
2316 | 2316 | self.n_prof_released += 1 |
|
|
2317 | 2317 | |
|
|
2318 | 2318 | if self.end_prof >= (self.n +self.lenProfileOut): |
|
|
2319 | 2319 | |
|
|
2320 | 2320 | self.init_prof = 0 |
|
|
2321 | 2321 | self.__profIndex = 0 |
|
|
2322 | 2322 | self.buffer = None |
|
|
2323 | 2323 | dataOut.buffer_empty = True |
|
|
2324 | 2324 | self.outliers_IDs_list = [] |
|
|
2325 | 2325 | self.n_prof_released = 0 |
|
|
2326 | 2326 | dataOut.flagNoData = False #enviar ultimo aunque sea outlier :( |
|
|
2327 | 2327 | #print("cleaning...", dataOut.buffer_empty) |
|
|
2328 | 2328 | dataOut.profileIndex = 0 #self.lenProfileOut |
|
|
2329 | 2329 | #################################################################### |
|
|
2330 | 2330 | return dataOut |
|
|
2331 | 2331 | |
|
|
2332 | 2332 | |
|
|
2333 | 2333 | #print("tp 223 ",dataOut.data.shape) |
|
|
2334 | 2334 | dataOut.flagNoData = True |
|
|
2335 | 2335 | |
|
|
2336 | 2336 | |
|
|
2337 | 2337 | |
|
|
2338 | 2338 | try: |
|
|
2339 | 2339 | #dataBlock = self.getData(dataOut.data.reshape(self.nChannels,1,self.nHeights), dataOut.utctime) |
|
|
2340 | 2340 | dataBlock = self.getData(numpy.reshape(dataOut.data,(self.nChannels,1,self.nHeights)), dataOut.utctime) |
|
|
2341 | 2341 | self.__count_exec +=1 |
|
|
2342 | 2342 | except Exception as e: |
|
|
2343 | 2343 | print("Error getting profiles data",self.__count_exec ) |
|
|
2344 | 2344 | print(e) |
|
|
2345 | 2345 | sys.exit() |
|
|
2346 | 2346 | |
|
|
2347 | 2347 | if self.__dataReady: |
|
|
2348 | 2348 | #print("omitting: ", len(self.outliers_IDs_list)) |
|
|
2349 | 2349 | self.__count_exec = 0 |
|
|
2350 | 2350 | #dataOut.data = |
|
|
2351 | 2351 | #self.buffer = numpy.flip(dataBlock, axis=1) |
|
|
2352 | 2352 | self.buffer = dataBlock |
|
|
2353 | 2353 | self.first_utcBlock = self.__initime |
|
|
2354 | 2354 | dataOut.utctime = self.__initime |
|
|
2355 | 2355 | dataOut.nProfiles = self.__profIndex |
|
|
2356 | 2356 | #dataOut.flagNoData = False |
|
|
2357 | 2357 | self.init_prof = 0 |
|
|
2358 | 2358 | self.__profIndex = 0 |
|
|
2359 | 2359 | self.__initime = None |
|
|
2360 | 2360 | dataBlock = None |
|
|
2361 | 2361 | self.__buffer_times = [] |
|
|
2362 | 2362 | dataOut.error = False |
|
|
2363 | 2363 | dataOut.useInputBuffer = True |
|
|
2364 | 2364 | dataOut.buffer_empty = False |
|
|
2365 | 2365 | #print("1 ch: {} prof: {} hs: {}".format(int(dataOut.nChannels),int(dataOut.nProfiles),int(dataOut.nHeights))) |
|
|
2366 | 2366 | |
|
|
2367 | 2367 | |
|
|
2368 | 2368 | |
|
|
2369 | 2369 | #print(self.__count_exec) |
|
|
2370 | 2370 | |
|
|
2371 | 2371 | return dataOut |
|
|
2372 | 2372 | |
|
|
2373 | 2373 | |
|
|
2374 | 2374 | class RemoveProfileSats2(Operation): |
|
|
2375 | 2375 | ''' |
|
|
2376 | 2376 | Escrito: Joab Apaza |
|
|
2377 | 2377 | |
|
|
2378 | 2378 | Omite los perfiles contaminados con seΓ±al de satΓ©lites, usando una altura de referencia |
|
|
2379 | 2379 | promedia todas las alturas para los cΓ‘lculos |
|
|
2380 | 2380 | In: |
|
|
2381 | 2381 | n = Cantidad de perfiles que se acumularan, usualmente 10 segundos |
|
|
2382 | 2382 | navg = Porcentaje de perfiles que puede considerarse como satΓ©lite, mΓ‘ximo 90% |
|
|
2383 | 2383 | minHei = |
|
|
2384 | 2384 | minRef = |
|
|
2385 | 2385 | maxRef = |
|
|
2386 | 2386 | nBins = |
|
|
2387 | 2387 | profile_margin = |
|
|
2388 | 2388 | th_hist_outlier = |
|
|
2389 | 2389 | nProfilesOut = |
|
|
2390 | 2390 | |
|
|
2391 | 2391 | Pensado para remover interferencias de las YAGI, se puede adaptar a otras interferencias |
|
|
2392 | 2392 | |
|
|
2393 | 2393 | remYagi = Activa la funcion de remociΓ³n de interferencias de la YAGI |
|
|
2394 | 2394 | nProfYagi = Cantidad de perfiles que son afectados, acorde NTX de la YAGI |
|
|
2395 | 2395 | offYagi = |
|
|
2396 | 2396 | minHJULIA = Altura mΓnima donde aparece la seΓ±al referencia de JULIA (-50) |
|
|
2397 | 2397 | maxHJULIA = Altura mΓ‘xima donde aparece la seΓ±al referencia de JULIA (-15) |
|
|
2398 | 2398 | |
|
|
2399 | 2399 | debug = Activa los grΓ‘ficos, recomendable ejecutar para ajustar los parΓ‘metros |
|
|
2400 | 2400 | para un experimento en especΓfico. |
|
|
2401 | 2401 | |
|
|
2402 | 2402 | ** se modifica para remover interferencias puntuales, es decir, desde otros radares. |
|
|
2403 | 2403 | Inicialmente se ha configurado para omitir tambiΓ©n los perfiles de la YAGI en los datos |
|
|
2404 | 2404 | de AMISR-ISR. |
|
|
2405 | 2405 | |
|
|
2406 | 2406 | Out: |
|
|
2407 | 2407 | profile clean |
|
|
2408 | 2408 | ''' |
|
|
2409 | 2409 | |
|
|
2410 | 2410 | |
|
|
2411 | 2411 | __buffer_data = [] |
|
|
2412 | 2412 | __buffer_times = [] |
|
|
2413 | 2413 | |
|
|
2414 | 2414 | buffer = None |
|
|
2415 | 2415 | |
|
|
2416 | 2416 | outliers_IDs_list = [] |
|
|
2417 | 2417 | |
|
|
2418 | 2418 | |
|
|
2419 | 2419 | __slots__ = ('n','navg','profileMargin','thHistOutlier','minHei_idx','maxHei_idx','nHeights', |
|
|
2420 | 2420 | 'first_utcBlock','__profIndex','init_prof','end_prof','lenProfileOut','nChannels','cohFactor', |
|
|
2421 | 2421 | '__count_exec','__initime','__dataReady','__ipp', 'minRef', 'maxRef', 'debug','prev_pnoise','thfactor') |
|
|
2422 | 2422 | def __init__(self, **kwargs): |
|
|
2423 | 2423 | |
|
|
2424 | 2424 | Operation.__init__(self, **kwargs) |
|
|
2425 | 2425 | self.isConfig = False |
|
|
2426 | 2426 | self.currentTime = None |
|
|
2427 | 2427 | |
|
|
2428 | 2428 | def setup(self,dataOut, n=None , navg=0.9, profileMargin=50,thHistOutlier=15,minHei=None, maxHei=None, nBins=10, |
|
|
2429 | 2429 | minRef=None, maxRef=None, debug=False, remYagi=False, nProfYagi = 0, offYagi=0, minHJULIA=None, maxHJULIA=None, |
|
|
2430 | 2430 | idate=None,startH=None,endH=None, thfactor=1 ): |
|
|
2431 | 2431 | |
|
|
2432 | 2432 | if n == None and timeInterval == None: |
|
|
2433 | 2433 | raise ValueError("nprofiles or timeInterval should be specified ...") |
|
|
2434 | 2434 | |
|
|
2435 | 2435 | if n != None: |
|
|
2436 | 2436 | self.n = n |
|
|
2437 | 2437 | |
|
|
2438 | 2438 | self.navg = navg |
|
|
2439 | 2439 | self.profileMargin = profileMargin |
|
|
2440 | 2440 | self.thHistOutlier = thHistOutlier |
|
|
2441 | 2441 | self.__profIndex = 0 |
|
|
2442 | 2442 | self.buffer = None |
|
|
2443 | 2443 | self._ipp = dataOut.ippSeconds |
|
|
2444 | 2444 | self.n_prof_released = 0 |
|
|
2445 | 2445 | self.heightList = dataOut.heightList |
|
|
2446 | 2446 | self.init_prof = 0 |
|
|
2447 | 2447 | self.end_prof = 0 |
|
|
2448 | 2448 | self.__count_exec = 0 |
|
|
2449 | 2449 | self.__profIndex = 0 |
|
|
2450 | 2450 | self.first_utcBlock = None |
|
|
2451 | 2451 | self.prev_pnoise = None |
|
|
2452 | 2452 | self.nBins = nBins |
|
|
2453 | 2453 | self.thfactor = thfactor |
|
|
2454 | 2454 | #self.__dh = dataOut.heightList[1] - dataOut.heightList[0] |
|
|
2455 | 2455 | minHei = minHei |
|
|
2456 | 2456 | maxHei = maxHei |
|
|
2457 | 2457 | if minHei==None : |
|
|
2458 | 2458 | minHei = dataOut.heightList[0] |
|
|
2459 | 2459 | if maxHei==None : |
|
|
2460 | 2460 | maxHei = dataOut.heightList[-1] |
|
|
2461 | 2461 | self.minHei_idx,self.maxHei_idx = getHei_index(minHei, maxHei, dataOut.heightList) |
|
|
2462 | 2462 | self.min_ref, self.max_ref = getHei_index(minRef, maxRef, dataOut.heightList) |
|
|
2463 | 2463 | self.nChannels = dataOut.nChannels |
|
|
2464 | 2464 | self.nHeights = dataOut.nHeights |
|
|
2465 | 2465 | self.test_counter = 0 |
|
|
2466 | 2466 | self.debug = debug |
|
|
2467 | 2467 | self.remYagi = remYagi |
|
|
2468 | 2468 | self.cohFactor = dataOut.nCohInt |
|
|
2469 | 2469 | if self.remYagi : |
|
|
2470 | 2470 | if minHJULIA==None or maxHJULIA==None: |
|
|
2471 | 2471 | raise ValueError("Parameters minHYagi and minHYagi are necessary!") |
|
|
2472 | 2472 | return |
|
|
2473 | 2473 | if idate==None or startH==None or endH==None: |
|
|
2474 | 2474 | raise ValueError("Date and hour parameters are necessary!") |
|
|
2475 | 2475 | return |
|
|
2476 | 2476 | self.minHJULIA_idx,self.maxHJULIA_idx = getHei_index(minHJULIA, maxHJULIA, dataOut.heightList) |
|
|
2477 | 2477 | self.offYagi = offYagi |
|
|
2478 | 2478 | self.nTxYagi = nProfYagi |
|
|
2479 | 2479 | |
|
|
2480 | 2480 | self.startTime = datetime.datetime.combine(idate,startH) |
|
|
2481 | 2481 | self.endTime = datetime.datetime.combine(idate,endH) |
|
|
2482 | 2482 | |
|
|
2483 | 2483 | log.warning("Be careful with the selection of parameters for sats removal! It is avisable to \ |
|
|
2484 | 2484 | activate the debug parameter in this operation for calibration", self.name) |
|
|
2485 | 2485 | |
|
|
2486 | 2486 | |
|
|
2487 | 2487 | def filterSatsProfiles(self): |
|
|
2488 | 2488 | |
|
|
2489 | 2489 | data = self.__buffer_data.copy() |
|
|
2490 | 2490 | #print(data.shape) |
|
|
2491 | 2491 | nChannels, profiles, heights = data.shape |
|
|
2492 | 2492 | indexes=numpy.zeros([], dtype=int) |
|
|
2493 | 2493 | indexes = numpy.delete(indexes,0) |
|
|
2494 | 2494 | |
|
|
2495 | 2495 | indexesYagi=numpy.zeros([], dtype=int) |
|
|
2496 | 2496 | indexesYagi = numpy.delete(indexesYagi,0) |
|
|
2497 | 2497 | |
|
|
2498 | 2498 | indexesYagi_up=numpy.zeros([], dtype=int) |
|
|
2499 | 2499 | indexesYagi_up = numpy.delete(indexesYagi_up,0) |
|
|
2500 | 2500 | indexesYagi_down=numpy.zeros([], dtype=int) |
|
|
2501 | 2501 | indexesYagi_down = numpy.delete(indexesYagi_down,0) |
|
|
2502 | 2502 | |
|
|
2503 | 2503 | |
|
|
2504 | 2504 | indexesJULIA=numpy.zeros([], dtype=int) |
|
|
2505 | 2505 | indexesJULIA = numpy.delete(indexesJULIA,0) |
|
|
2506 | 2506 | |
|
|
2507 | 2507 | outliers_IDs=[] |
|
|
2508 | 2508 | |
|
|
2509 | 2509 | div = profiles//self.nBins |
|
|
2510 | 2510 | |
|
|
2511 | 2511 | for c in range(nChannels): |
|
|
2512 | 2512 | #print(self.min_ref,self.max_ref) |
|
|
2513 | 2513 | |
|
|
2514 | 2514 | import scipy.signal |
|
|
2515 | 2515 | b, a = scipy.signal.butter(3, 0.5) |
|
|
2516 | 2516 | #noise_ref = (data[c,:,self.min_ref:self.max_ref] * numpy.conjugate(data[c,:,self.min_ref:self.max_ref])) |
|
|
2517 | 2517 | noise_ref = numpy.abs(data[c,:,self.min_ref:self.max_ref]) |
|
|
2518 | 2518 | lnoise = len(noise_ref[0,:]) |
|
|
2519 | 2519 | #print(noise_ref.shape) |
|
|
2520 | 2520 | noise_ref = noise_ref.mean(axis=1) |
|
|
2521 | 2521 | #fnoise = noise_ref |
|
|
2522 | 2522 | fnoise = scipy.signal.filtfilt(b, a, noise_ref) |
|
|
2523 | 2523 | #noise_refdB = 10* numpy.log10(noise_ref) |
|
|
2524 | 2524 | #print("Noise ",numpy.percentile(noise_ref,95)) |
|
|
2525 | 2525 | p95 = numpy.percentile(fnoise,95) |
|
|
2526 | 2526 | mean_noise = fnoise.mean() |
|
|
2527 | 2527 | |
|
|
2528 | 2528 | if self.prev_pnoise != None: |
|
|
2529 | 2529 | if mean_noise < (1.1 * self.prev_pnoise) and mean_noise > (0.9 * self.prev_pnoise): |
|
|
2530 | 2530 | mean_noise = 0.9*mean_noise + 0.1*self.prev_pnoise |
|
|
2531 | 2531 | self.prev_pnoise = mean_noise |
|
|
2532 | 2532 | else: |
|
|
2533 | 2533 | mean_noise = self.prev_pnoise |
|
|
2534 | 2534 | else: |
|
|
2535 | 2535 | self.prev_pnoise = mean_noise |
|
|
2536 | 2536 | |
|
|
2537 | 2537 | std = fnoise.std()+ fnoise.mean() |
|
|
2538 | 2538 | |
|
|
2539 | 2539 | |
|
|
2540 | 2540 | |
|
|
2541 | 2541 | #power = (data[c,:,self.minHei_idx:self.maxHei_idx] * numpy.conjugate(data[c,:,self.minHei_idx:self.maxHei_idx])) |
|
|
2542 | 2542 | power = numpy.abs(data[c,:,self.minHei_idx:self.maxHei_idx]) |
|
|
2543 | 2543 | npower = len(power[0,:]) |
|
|
2544 | 2544 | #print(power.shape) |
|
|
2545 | 2545 | power = power.mean(axis=1) |
|
|
2546 | 2546 | |
|
|
2547 | 2547 | fpower = scipy.signal.filtfilt(b, a, power) |
|
|
2548 | 2548 | #print(power.shape) |
|
|
2549 | 2549 | #powerdB = 10* numpy.log10(power) |
|
|
2550 | 2550 | |
|
|
2551 | 2551 | #th = p95 * self.thfactor |
|
|
2552 | 2552 | th = mean_noise * self.thfactor |
|
|
2553 | 2553 | |
|
|
2554 | 2554 | index = numpy.where(fpower > th ) |
|
|
2555 | 2555 | #print("Noise ",mean_noise, p95) |
|
|
2556 | 2556 | #print(index) |
|
|
2557 | 2557 | |
|
|
2558 | 2558 | |
|
|
2559 | 2559 | |
|
|
2560 | 2560 | if index[0].size <= int(self.navg*profiles): #outliers from sats |
|
|
2561 | 2561 | indexes = numpy.append(indexes, index[0]) |
|
|
2562 | 2562 | |
|
|
2563 | 2563 | index2low = numpy.where(fpower < (th*0.5 )) #outliers from no TX |
|
|
2564 | 2564 | if index2low[0].size <= int(self.navg*profiles): |
|
|
2565 | 2565 | indexes = numpy.append(indexes, index2low[0]) |
|
|
2566 | 2566 | |
|
|
2567 | 2567 | #print("sdas ", noise_ref.mean()) |
|
|
2568 | 2568 | |
|
|
2569 | 2569 | if self.remYagi : |
|
|
2570 | 2570 | #print(self.minHJULIA_idx, self.maxHJULIA_idx) |
|
|
2571 | 2571 | powerJULIA = (data[c,:,self.minHJULIA_idx:self.maxHJULIA_idx] * numpy.conjugate(data[c,:,self.minHJULIA_idx:self.maxHJULIA_idx])).real |
|
|
2572 | 2572 | powerJULIA = powerJULIA.mean(axis=1) |
|
|
2573 | 2573 | th_JULIA = powerJULIA.mean()*0.85 |
|
|
2574 | 2574 | indexJULIA = numpy.where(powerJULIA >= th_JULIA ) |
|
|
2575 | 2575 | |
|
|
2576 | 2576 | indexesJULIA= numpy.append(indexesJULIA, indexJULIA[0]) |
|
|
2577 | 2577 | |
|
|
2578 | 2578 | # fig, ax = plt.subplots() |
|
|
2579 | 2579 | # ax.plot(powerJULIA) |
|
|
2580 | 2580 | # ax.axhline(th_JULIA, color='r') |
|
|
2581 | 2581 | # plt.grid() |
|
|
2582 | 2582 | # plt.show() |
|
|
2583 | 2583 | |
|
|
2584 | 2584 | if self.debug: |
|
|
2585 | 2585 | fig, ax = plt.subplots() |
|
|
2586 | 2586 | ax.plot(fpower, label="power") |
|
|
2587 | 2587 | #ax.plot(fnoise, label="noise ref") |
|
|
2588 | 2588 | ax.axhline(th, color='g', label="th") |
|
|
2589 | 2589 | #ax.axhline(std, color='b', label="mean") |
|
|
2590 | 2590 | ax.legend() |
|
|
2591 | 2591 | plt.grid() |
|
|
2592 | 2592 | plt.show() |
|
|
2593 | 2593 | |
|
|
2594 | 2594 | #print(indexes) |
|
|
2595 | 2595 | |
|
|
2596 | 2596 | #outliers_IDs = outliers_IDs.astype(numpy.dtype('int64')) |
|
|
2597 | 2597 | #outliers_IDs = numpy.unique(outliers_IDs) |
|
|
2598 | 2598 | # print(indexesJULIA) |
|
|
2599 | 2599 | if len(indexesJULIA > 1): |
|
|
2600 | 2600 | iJ = indexesJULIA |
|
|
2601 | 2601 | locs = [ (iJ[n]-iJ[n-1]) > 5 for n in range(len(iJ))] |
|
|
2602 | 2602 | locs_2 = numpy.where(locs)[0] |
|
|
2603 | 2603 | #print(locs_2, indexesJULIA[locs_2-1]) |
|
|
2604 | 2604 | indexesYagi_up = numpy.append(indexesYagi_up, indexesJULIA[locs_2-1]) |
|
|
2605 | 2605 | indexesYagi_down = numpy.append(indexesYagi_down, indexesJULIA[locs_2]) |
|
|
2606 | 2606 | |
|
|
2607 | 2607 | |
|
|
2608 | 2608 | indexesYagi_up = numpy.append(indexesYagi_up,indexesJULIA[-1]) |
|
|
2609 | 2609 | indexesYagi_down = numpy.append(indexesYagi_down,indexesJULIA[0]) |
|
|
2610 | 2610 | |
|
|
2611 | 2611 | indexesYagi_up = numpy.unique(indexesYagi_up) |
|
|
2612 | 2612 | indexesYagi_down = numpy.unique(indexesYagi_down) |
|
|
2613 | 2613 | |
|
|
2614 | 2614 | |
|
|
2615 | 2615 | aux_ind = [ numpy.arange( (self.offYagi + k)+1, (self.offYagi + k + self.nTxYagi)+1, 1, dtype=int) for k in indexesYagi_up] |
|
|
2616 | 2616 | indexesYagi_up = (numpy.asarray(aux_ind)).flatten() |
|
|
2617 | 2617 | |
|
|
2618 | 2618 | aux_ind2 = [ numpy.arange( (k - self.nTxYagi)+1, k+1 , 1, dtype=int) for k in indexesYagi_down] |
|
|
2619 | 2619 | indexesYagi_down = (numpy.asarray(aux_ind2)).flatten() |
|
|
2620 | 2620 | |
|
|
2621 | 2621 | indexesYagi = numpy.append(indexesYagi,indexesYagi_up) |
|
|
2622 | 2622 | indexesYagi = numpy.append(indexesYagi,indexesYagi_down) |
|
|
2623 | 2623 | |
|
|
2624 | 2624 | |
|
|
2625 | 2625 | indexesYagi = indexesYagi[ (indexesYagi >= 0) & (indexesYagi<profiles)] |
|
|
2626 | 2626 | indexesYagi = numpy.unique(indexesYagi) |
|
|
2627 | 2627 | |
|
|
2628 | 2628 | #print("indexes: " ,indexes) |
|
|
2629 | 2629 | outs_lines = numpy.unique(indexes) |
|
|
2630 | 2630 | #print(outs_lines) |
|
|
2631 | 2631 | |
|
|
2632 | 2632 | #Agrupando el histograma de outliers, |
|
|
2633 | 2633 | my_bins = numpy.linspace(0,int(profiles), div, endpoint=True) |
|
|
2634 | 2634 | |
|
|
2635 | 2635 | |
|
|
2636 | 2636 | hist, bins = numpy.histogram(outs_lines,bins=my_bins) |
|
|
2637 | 2637 | #print("hist: ",hist) |
|
|
2638 | 2638 | hist_outliers_indexes = numpy.where(hist >= self.thHistOutlier)[0] #es outlier |
|
|
2639 | 2639 | # print(hist_outliers_indexes) |
|
|
2640 | 2640 | if len(hist_outliers_indexes>0): |
|
|
2641 | 2641 | hist_outliers_indexes = numpy.append(hist_outliers_indexes,hist_outliers_indexes[-1]+1) |
|
|
2642 | 2642 | |
|
|
2643 | 2643 | bins_outliers_indexes = [int(i)+1 for i in (bins[hist_outliers_indexes])] # |
|
|
2644 | 2644 | outlier_loc_index = [] |
|
|
2645 | 2645 | #print("out indexes ", bins_outliers_indexes) |
|
|
2646 | 2646 | |
|
|
2647 | 2647 | # if len(bins_outliers_indexes) <= 2: |
|
|
2648 | 2648 | # extprof = 0 |
|
|
2649 | 2649 | # else: |
|
|
2650 | 2650 | # extprof = self.profileMargin |
|
|
2651 | 2651 | |
|
|
2652 | 2652 | extprof = self.profileMargin |
|
|
2653 | 2653 | |
|
|
2654 | 2654 | outlier_loc_index = [e for n in range(len(bins_outliers_indexes)) for e in range(bins_outliers_indexes[n]-extprof,bins_outliers_indexes[n] + extprof) ] |
|
|
2655 | 2655 | outlier_loc_index = numpy.asarray(outlier_loc_index) |
|
|
2656 | 2656 | # if len(outlier_loc_index)>1: |
|
|
2657 | 2657 | # ipmax = numpy.where(fpower==fpower.max())[0] |
|
|
2658 | 2658 | # print("pmax: ",ipmax) |
|
|
2659 | 2659 | |
|
|
2660 | 2660 | |
|
|
2661 | 2661 | |
|
|
2662 | 2662 | |
|
|
2663 | 2663 | #print("outliers Ids: ", outlier_loc_index, outlier_loc_index.shape) |
|
|
2664 | 2664 | outlier_loc_index = outlier_loc_index[ (outlier_loc_index >= 0) & (outlier_loc_index<profiles)] |
|
|
2665 | 2665 | #print("outliers final: ", outlier_loc_index) |
|
|
2666 | 2666 | |
|
|
2667 | 2667 | |
|
|
2668 | 2668 | if self.debug: |
|
|
2669 | 2669 | x, y = numpy.meshgrid(numpy.arange(profiles), self.heightList) |
|
|
2670 | 2670 | fig, ax = plt.subplots(nChannels,2,figsize=(8, 6)) |
|
|
2671 | 2671 | |
|
|
2672 | 2672 | for i in range(nChannels): |
|
|
2673 | 2673 | dat = data[i,:,:].real |
|
|
2674 | 2674 | dat = 10* numpy.log10((data[i,:,:] * numpy.conjugate(data[i,:,:])).real) |
|
|
2675 | 2675 | m = numpy.nanmean(dat) |
|
|
2676 | 2676 | o = numpy.nanstd(dat) |
|
|
2677 | 2677 | if nChannels>1: |
|
|
2678 | 2678 | c = ax[i][0].pcolormesh(x, y, dat.T, cmap ='jet', vmin = 60, vmax = 70) |
|
|
2679 | 2679 | ax[i][0].vlines(outs_lines,650,700, linestyles='dashed', label = 'outs', color='w') |
|
|
2680 | 2680 | #fig.colorbar(c) |
|
|
2681 | 2681 | ax[i][0].vlines(outlier_loc_index,700,750, linestyles='dashed', label = 'outs', color='r') |
|
|
2682 | 2682 | ax[i][1].hist(outs_lines,bins=my_bins) |
|
|
2683 | 2683 | if self.remYagi : |
|
|
2684 | 2684 | ax[0].vlines(indexesYagi,750,850, linestyles='dashed', label = 'yagi', color='m') |
|
|
2685 | 2685 | else: |
|
|
2686 | 2686 | c = ax[0].pcolormesh(x, y, dat.T, cmap ='jet', vmin = 60, vmax = (70+2*self.cohFactor)) |
|
|
2687 | 2687 | ax[0].vlines(outs_lines,650,700, linestyles='dashed', label = 'outs', color='w') |
|
|
2688 | 2688 | #fig.colorbar(c) |
|
|
2689 | 2689 | ax[0].vlines(outlier_loc_index,700,750, linestyles='dashed', label = 'outs', color='r') |
|
|
2690 | 2690 | |
|
|
2691 | 2691 | ax[1].hist(outs_lines,bins=my_bins) |
|
|
2692 | 2692 | if self.remYagi : |
|
|
2693 | 2693 | ax[0].vlines(indexesYagi,750,850, linestyles='dashed', label = 'yagi', color='m') |
|
|
2694 | 2694 | plt.show() |
|
|
2695 | 2695 | |
|
|
2696 | 2696 | |
|
|
2697 | 2697 | |
|
|
2698 | 2698 | |
|
|
2699 | 2699 | if self.remYagi and (self.currentTime < self.startTime and self.currentTime < self.endTime): |
|
|
2700 | 2700 | outlier_loc_index = numpy.append(outlier_loc_index,indexesYagi) |
|
|
2701 | 2701 | |
|
|
2702 | 2702 | self.outliers_IDs_list = numpy.unique(outlier_loc_index) |
|
|
2703 | 2703 | |
|
|
2704 | 2704 | #print("outs list: ", self.outliers_IDs_list) |
|
|
2705 | 2705 | return self.__buffer_data |
|
|
2706 | 2706 | |
|
|
2707 | 2707 | |
|
|
2708 | 2708 | |
|
|
2709 | 2709 | def fillBuffer(self, data, datatime): |
|
|
2710 | 2710 | |
|
|
2711 | 2711 | if self.__profIndex == 0: |
|
|
2712 | 2712 | self.__buffer_data = data.copy() |
|
|
2713 | 2713 | |
|
|
2714 | 2714 | else: |
|
|
2715 | 2715 | self.__buffer_data = numpy.concatenate((self.__buffer_data,data), axis=1)#en perfiles |
|
|
2716 | 2716 | self.__profIndex += 1 |
|
|
2717 | 2717 | self.__buffer_times.append(datatime) |
|
|
2718 | 2718 | |
|
|
2719 | 2719 | def getData(self, data, datatime=None): |
|
|
2720 | 2720 | |
|
|
2721 | 2721 | if self.__profIndex == 0: |
|
|
2722 | 2722 | self.__initime = datatime |
|
|
2723 | 2723 | |
|
|
2724 | 2724 | |
|
|
2725 | 2725 | self.__dataReady = False |
|
|
2726 | 2726 | |
|
|
2727 | 2727 | self.fillBuffer(data, datatime) |
|
|
2728 | 2728 | dataBlock = None |
|
|
2729 | 2729 | |
|
|
2730 | 2730 | if self.__profIndex == self.n: |
|
|
2731 | 2731 | #print("apnd : ",data) |
|
|
2732 | 2732 | dataBlock = self.filterSatsProfiles() |
|
|
2733 | 2733 | self.__dataReady = True |
|
|
2734 | 2734 | |
|
|
2735 | 2735 | return dataBlock |
|
|
2736 | 2736 | |
|
|
2737 | 2737 | if dataBlock is None: |
|
|
2738 | 2738 | return None, None |
|
|
2739 | 2739 | |
|
|
2740 | 2740 | |
|
|
2741 | 2741 | |
|
|
2742 | 2742 | return dataBlock |
|
|
2743 | 2743 | |
|
|
2744 | 2744 | def releaseBlock(self): |
|
|
2745 | 2745 | |
|
|
2746 | 2746 | if self.n % self.lenProfileOut != 0: |
|
|
2747 | 2747 | raise ValueError("lenProfileOut %d must be submultiple of nProfiles %d" %(self.lenProfileOut, self.n)) |
|
|
2748 | 2748 | return None |
|
|
2749 | 2749 | |
|
|
2750 | 2750 | data = self.buffer[:,self.init_prof:self.end_prof:,:] #ch, prof, alt |
|
|
2751 | 2751 | |
|
|
2752 | 2752 | self.init_prof = self.end_prof |
|
|
2753 | 2753 | self.end_prof += self.lenProfileOut |
|
|
2754 | 2754 | #print("data release shape: ",dataOut.data.shape, self.end_prof) |
|
|
2755 | 2755 | self.n_prof_released += 1 |
|
|
2756 | 2756 | |
|
|
2757 | 2757 | return data |
|
|
2758 | 2758 | |
|
|
2759 | 2759 | def run(self, dataOut, n=None, navg=0.9, nProfilesOut=1, profile_margin=50, th_hist_outlier=15,minHei=None,nBins=10, |
|
|
2760 | 2760 | maxHei=None, minRef=None, maxRef=None, debug=False, remYagi=False, nProfYagi = 0, offYagi=0, minHJULIA=None, maxHJULIA=None, |
|
|
2761 | 2761 | idate=None,startH=None,endH=None, thfactor=1): |
|
|
2762 | 2762 | |
|
|
2763 | 2763 | if not self.isConfig: |
|
|
2764 | 2764 | #print("init p idx: ", dataOut.profileIndex ) |
|
|
2765 | 2765 | self.setup(dataOut,n=n, navg=navg,profileMargin=profile_margin,thHistOutlier=th_hist_outlier,minHei=minHei, |
|
|
2766 | 2766 | nBins=10, maxHei=maxHei, minRef=minRef, maxRef=maxRef, debug=debug, remYagi=remYagi, nProfYagi = nProfYagi, |
|
|
2767 | 2767 | offYagi=offYagi, minHJULIA=minHJULIA,maxHJULIA=maxHJULIA,idate=idate,startH=startH,endH=endH, thfactor=thfactor) |
|
|
2768 | 2768 | |
|
|
2769 | 2769 | self.isConfig = True |
|
|
2770 | 2770 | |
|
|
2771 | 2771 | dataBlock = None |
|
|
2772 | 2772 | self.currentTime = datetime.datetime.fromtimestamp(dataOut.utctime) |
|
|
2773 | 2773 | |
|
|
2774 | 2774 | if not dataOut.buffer_empty: #hay datos acumulados |
|
|
2775 | 2775 | |
|
|
2776 | 2776 | if self.init_prof == 0: |
|
|
2777 | 2777 | self.n_prof_released = 0 |
|
|
2778 | 2778 | self.lenProfileOut = nProfilesOut |
|
|
2779 | 2779 | dataOut.flagNoData = False |
|
|
2780 | 2780 | #print("tp 2 ",dataOut.data.shape) |
|
|
2781 | 2781 | |
|
|
2782 | 2782 | self.init_prof = 0 |
|
|
2783 | 2783 | self.end_prof = self.lenProfileOut |
|
|
2784 | 2784 | |
|
|
2785 | 2785 | dataOut.nProfiles = self.lenProfileOut |
|
|
2786 | 2786 | if nProfilesOut == 1: |
|
|
2787 | 2787 | dataOut.flagDataAsBlock = False |
|
|
2788 | 2788 | else: |
|
|
2789 | 2789 | dataOut.flagDataAsBlock = True |
|
|
2790 | 2790 | #print("prof: ",self.init_prof) |
|
|
2791 | 2791 | dataOut.flagNoData = False |
|
|
2792 | 2792 | if numpy.isin(self.n_prof_released, self.outliers_IDs_list): |
|
|
2793 | 2793 | #print("omitting: ", self.n_prof_released) |
|
|
2794 | 2794 | dataOut.flagNoData = True |
|
|
2795 | 2795 | dataOut.ippSeconds = self._ipp |
|
|
2796 | 2796 | dataOut.utctime = self.first_utcBlock + self.init_prof*self._ipp |
|
|
2797 | 2797 | # print("time: ", dataOut.utctime, self.first_utcBlock, self.init_prof,self._ipp,dataOut.ippSeconds) |
|
|
2798 | 2798 | #dataOut.data = self.releaseBlock() |
|
|
2799 | 2799 | #########################################################3 |
|
|
2800 | 2800 | if self.n % self.lenProfileOut != 0: |
|
|
2801 | 2801 | raise ValueError("lenProfileOut %d must be submultiple of nProfiles %d" %(self.lenProfileOut, self.n)) |
|
|
2802 | 2802 | return None |
|
|
2803 | 2803 | |
|
|
2804 | 2804 | dataOut.data = None |
|
|
2805 | 2805 | |
|
|
2806 | 2806 | if nProfilesOut == 1: |
|
|
2807 | 2807 | dataOut.data = self.buffer[:,self.end_prof-1,:] #ch, prof, alt |
|
|
2808 | 2808 | else: |
|
|
2809 | 2809 | dataOut.data = self.buffer[:,self.init_prof:self.end_prof,:] #ch, prof, alt |
|
|
2810 | 2810 | |
|
|
2811 | 2811 | self.init_prof = self.end_prof |
|
|
2812 | 2812 | self.end_prof += self.lenProfileOut |
|
|
2813 | 2813 | #print("data release shape: ",dataOut.data.shape, self.end_prof, dataOut.flagNoData) |
|
|
2814 | 2814 | self.n_prof_released += 1 |
|
|
2815 | 2815 | |
|
|
2816 | 2816 | if self.end_prof >= (self.n +self.lenProfileOut): |
|
|
2817 | 2817 | |
|
|
2818 | 2818 | self.init_prof = 0 |
|
|
2819 | 2819 | self.__profIndex = 0 |
|
|
2820 | 2820 | self.buffer = None |
|
|
2821 | 2821 | dataOut.buffer_empty = True |
|
|
2822 | 2822 | self.outliers_IDs_list = [] |
|
|
2823 | 2823 | self.n_prof_released = 0 |
|
|
2824 | 2824 | dataOut.flagNoData = False #enviar ultimo aunque sea outlier :( |
|
|
2825 | 2825 | #print("cleaning...", dataOut.buffer_empty) |
|
|
2826 | 2826 | dataOut.profileIndex = self.__profIndex |
|
|
2827 | 2827 | #################################################################### |
|
|
2828 | 2828 | return dataOut |
|
|
2829 | 2829 | |
|
|
2830 | 2830 | |
|
|
2831 | 2831 | #print("tp 223 ",dataOut.data.shape) |
|
|
2832 | 2832 | dataOut.flagNoData = True |
|
|
2833 | 2833 | |
|
|
2834 | 2834 | |
|
|
2835 | 2835 | |
|
|
2836 | 2836 | try: |
|
|
2837 | 2837 | #dataBlock = self.getData(dataOut.data.reshape(self.nChannels,1,self.nHeights), dataOut.utctime) |
|
|
2838 | 2838 | dataBlock = self.getData(numpy.reshape(dataOut.data,(self.nChannels,1,self.nHeights)), dataOut.utctime) |
|
|
2839 | 2839 | self.__count_exec +=1 |
|
|
2840 | 2840 | except Exception as e: |
|
|
2841 | 2841 | print("Error getting profiles data",self.__count_exec ) |
|
|
2842 | 2842 | print(e) |
|
|
2843 | 2843 | sys.exit() |
|
|
2844 | 2844 | |
|
|
2845 | 2845 | if self.__dataReady: |
|
|
2846 | 2846 | #print("omitting: ", len(self.outliers_IDs_list)) |
|
|
2847 | 2847 | self.__count_exec = 0 |
|
|
2848 | 2848 | #dataOut.data = |
|
|
2849 | 2849 | #self.buffer = numpy.flip(dataBlock, axis=1) |
|
|
2850 | 2850 | self.buffer = dataBlock |
|
|
2851 | 2851 | self.first_utcBlock = self.__initime |
|
|
2852 | 2852 | dataOut.utctime = self.__initime |
|
|
2853 | 2853 | dataOut.nProfiles = self.__profIndex |
|
|
2854 | 2854 | #dataOut.flagNoData = False |
|
|
2855 | 2855 | self.init_prof = 0 |
|
|
2856 | 2856 | self.__profIndex = 0 |
|
|
2857 | 2857 | self.__initime = None |
|
|
2858 | 2858 | dataBlock = None |
|
|
2859 | 2859 | self.__buffer_times = [] |
|
|
2860 | 2860 | dataOut.error = False |
|
|
2861 | 2861 | dataOut.useInputBuffer = True |
|
|
2862 | 2862 | dataOut.buffer_empty = False |
|
|
2863 | 2863 | #print("1 ch: {} prof: {} hs: {}".format(int(dataOut.nChannels),int(dataOut.nProfiles),int(dataOut.nHeights))) |
|
|
2864 | 2864 | |
|
|
2865 | 2865 | |
|
|
2866 | 2866 | |
|
|
2867 | 2867 | #print(self.__count_exec) |
|
|
2868 | 2868 | |
|
|
2869 | 2869 | return dataOut |
|
|
2870 | 2870 | |
|
|
2871 | 2871 | |
|
|
2872 | 2872 | |
|
|
2873 | 2873 | |
|
|
2874 | 2874 | class remHeightsIppInterf(Operation): |
|
|
2875 | 2875 | |
|
|
2876 | 2876 | def __init__(self, **kwargs): |
|
|
2877 | 2877 | |
|
|
2878 | 2878 | |
|
|
2879 | 2879 | Operation.__init__(self, **kwargs) |
|
|
2880 | 2880 | |
|
|
2881 | 2881 | self.isConfig = False |
|
|
2882 | 2882 | |
|
|
2883 | 2883 | self.heights_indx = None |
|
|
2884 | 2884 | self.heightsList = [] |
|
|
2885 | 2885 | |
|
|
2886 | 2886 | self.ipp1 = None |
|
|
2887 | 2887 | self.ipp2 = None |
|
|
2888 | 2888 | self.tx1 = None |
|
|
2889 | 2889 | self.tx2 = None |
|
|
2890 | 2890 | self.dh1 = None |
|
|
2891 | 2891 | |
|
|
2892 | 2892 | |
|
|
2893 | 2893 | def setup(self, dataOut, ipp1=None, ipp2=None, tx1=None, tx2=None, dh1=None, |
|
|
2894 | 2894 | idate=None, startH=None, endH=None): |
|
|
2895 | 2895 | |
|
|
2896 | 2896 | |
|
|
2897 | 2897 | self.ipp1 = ipp1 |
|
|
2898 | 2898 | self.ipp2 = ipp2 |
|
|
2899 | 2899 | self.tx1 = tx1 |
|
|
2900 | 2900 | self.tx2 = tx2 |
|
|
2901 | 2901 | self.dh1 = dh1 |
|
|
2902 | 2902 | |
|
|
2903 | 2903 | _maxIpp1R = dataOut.heightList.max() |
|
|
2904 | 2904 | |
|
|
2905 | 2905 | _n_repeats = int(_maxIpp1R / ipp2) |
|
|
2906 | 2906 | _init_hIntf = (tx1 + ipp2/2)+ dh1 |
|
|
2907 | 2907 | _n_hIntf = int(tx2 / dh1) |
|
|
2908 | 2908 | |
|
|
2909 | 2909 | self.heightsList = [_init_hIntf+n*ipp2 for n in range(_n_repeats) ] |
|
|
2910 | 2910 | heiList = dataOut.heightList |
|
|
2911 | 2911 | self.heights_indx = [getHei_index(h,h,heiList)[0] for h in self.heightsList] |
|
|
2912 | 2912 | |
|
|
2913 | 2913 | self.heights_indx = [ numpy.asarray([k for k in range(_n_hIntf+2)])+(getHei_index(h,h,heiList)[0] -1) for h in self.heightsList] |
|
|
2914 | 2914 | |
|
|
2915 | 2915 | self.heights_indx = numpy.asarray(self.heights_indx ) |
|
|
2916 | 2916 | self.isConfig = True |
|
|
2917 | 2917 | self.startTime = datetime.datetime.combine(idate,startH) |
|
|
2918 | 2918 | self.endTime = datetime.datetime.combine(idate,endH) |
|
|
2919 | 2919 | #print(self.startTime, self.endTime) |
|
|
2920 | 2920 | #print("nrepeats: ", _n_repeats, " _nH: ",_n_hIntf ) |
|
|
2921 | 2921 | |
|
|
2922 | 2922 | log.warning("Heights set to zero (km): ", self.name) |
|
|
2923 | 2923 | log.warning(str((dataOut.heightList[self.heights_indx].flatten())), self.name) |
|
|
2924 | 2924 | log.warning("Be careful with the selection of heights for noise calculation!") |
|
|
2925 | 2925 | |
|
|
2926 | 2926 | def run(self, dataOut, ipp1=None, ipp2=None, tx1=None, tx2=None, dh1=None, idate=None, |
|
|
2927 | 2927 | startH=None, endH=None): |
|
|
2928 | 2928 | #print(locals().values()) |
|
|
2929 | 2929 | if None in locals().values(): |
|
|
2930 | 2930 | log.warning('Missing kwargs, invalid values """None""" ', self.name) |
|
|
2931 | 2931 | return dataOut |
|
|
2932 | 2932 | |
|
|
2933 | 2933 | |
|
|
2934 | 2934 | if not self.isConfig: |
|
|
2935 | 2935 | self.setup(dataOut, ipp1=ipp1, ipp2=ipp2, tx1=tx1, tx2=tx2, dh1=dh1, |
|
|
2936 | 2936 | idate=idate, startH=startH, endH=endH) |
|
|
2937 | 2937 | |
|
|
2938 | 2938 | dataOut.flagProfilesByRange = False |
|
|
2939 | 2939 | currentTime = datetime.datetime.fromtimestamp(dataOut.utctime) |
|
|
2940 | 2940 | |
|
|
2941 | 2941 | if currentTime < self.startTime or currentTime > self.endTime: |
|
|
2942 | 2942 | return dataOut |
|
|
2943 | 2943 | |
|
|
2944 | 2944 | for ch in range(dataOut.data.shape[0]): |
|
|
2945 | 2945 | |
|
|
2946 | 2946 | for hk in self.heights_indx.flatten(): |
|
|
2947 | 2947 | if dataOut.data.ndim < 3: |
|
|
2948 | 2948 | dataOut.data[ch,hk] = 0.0 + 0.0j |
|
|
2949 | 2949 | else: |
|
|
2950 | 2950 | dataOut.data[ch,:,hk] = 0.0 + 0.0j |
|
|
2951 | 2951 | |
|
|
2952 | 2952 | dataOut.flagProfilesByRange = True |
|
|
2953 | 2953 | |
|
|
2954 | 2954 | return dataOut |
|
|
2955 | 2955 | |
|
|
2956 | 2956 | |
|
|
2957 | 2957 | |
|
|
2958 | 2958 | |
|
|
2959 | 2959 | class profiles2Block(Operation): |
|
|
2960 | 2960 | ''' |
|
|
2961 | 2961 | Escrito: Joab Apaza |
|
|
2962 | 2962 | |
|
|
2963 | 2963 | genera un bloque de perfiles |
|
|
2964 | 2964 | |
|
|
2965 | 2965 | |
|
|
2966 | 2966 | Out: |
|
|
2967 | 2967 | block |
|
|
2968 | 2968 | ''' |
|
|
2969 | 2969 | |
|
|
2970 | 2970 | isConfig = False |
|
|
2971 | 2971 | __buffer_data = [] |
|
|
2972 | 2972 | __buffer_times = [] |
|
|
2973 | 2973 | __profIndex = 0 |
|
|
2974 | 2974 | __byTime = False |
|
|
2975 | 2975 | __initime = None |
|
|
2976 | 2976 | __lastdatatime = None |
|
|
2977 | 2977 | buffer = None |
|
|
2978 | 2978 | n = None |
|
|
2979 | 2979 | __dataReady = False |
|
|
2980 | 2980 | __nChannels = None |
|
|
2981 | 2981 | __nHeis = None |
|
|
2982 | 2982 | |
|
|
2983 | 2983 | def __init__(self, **kwargs): |
|
|
2984 | 2984 | |
|
|
2985 | 2985 | Operation.__init__(self, **kwargs) |
|
|
2986 | 2986 | self.isConfig = False |
|
|
2987 | 2987 | |
|
|
2988 | 2988 | def setup(self,n=None, timeInterval=None): |
|
|
2989 | 2989 | |
|
|
2990 | 2990 | if n == None and timeInterval == None: |
|
|
2991 | 2991 | raise ValueError("n or timeInterval should be specified ...") |
|
|
2992 | 2992 | |
|
|
2993 | 2993 | if n != None: |
|
|
2994 | 2994 | self.n = n |
|
|
2995 | 2995 | self.__byTime = False |
|
|
2996 | 2996 | else: |
|
|
2997 | 2997 | self.__integrationtime = timeInterval #* 60. #if (type(timeInterval)!=integer) -> change this line |
|
|
2998 | 2998 | self.n = 9999 |
|
|
2999 | 2999 | self.__byTime = True |
|
|
3000 | 3000 | |
|
|
3001 | 3001 | self.__profIndex = 0 |
|
|
3002 | 3002 | |
|
|
3003 | 3003 | |
|
|
3004 | 3004 | def fillBuffer(self, data, datatime): |
|
|
3005 | 3005 | |
|
|
3006 | 3006 | if self.__profIndex == 0: |
|
|
3007 | 3007 | self.__buffer_data = data.copy() |
|
|
3008 | 3008 | |
|
|
3009 | 3009 | else: |
|
|
3010 | 3010 | self.__buffer_data = numpy.concatenate((self.__buffer_data,data), axis=1)#en perfiles |
|
|
3011 | 3011 | self.__profIndex += 1 |
|
|
3012 | 3012 | self.__buffer_times.append(datatime) |
|
|
3013 | 3013 | |
|
|
3014 | 3014 | def getData(self, data, datatime=None): |
|
|
3015 | 3015 | if self.__initime == None: |
|
|
3016 | 3016 | self.__initime = datatime |
|
|
3017 | 3017 | |
|
|
3018 | 3018 | if data.ndim < 3: |
|
|
3019 | 3019 | data = data.reshape(self.__nChannels,1,self.__nHeis ) |
|
|
3020 | 3020 | |
|
|
3021 | 3021 | if self.__byTime: |
|
|
3022 | 3022 | dataBlock = self.byTime(data, datatime) |
|
|
3023 | 3023 | else: |
|
|
3024 | 3024 | dataBlock = self.byProfiles(data, datatime) |
|
|
3025 | 3025 | |
|
|
3026 | 3026 | |
|
|
3027 | 3027 | self.__lastdatatime = datatime |
|
|
3028 | 3028 | |
|
|
3029 | 3029 | if dataBlock is None: |
|
|
3030 | 3030 | return None, None |
|
|
3031 | 3031 | |
|
|
3032 | 3032 | return dataBlock, self.__buffer_times |
|
|
3033 | 3033 | |
|
|
3034 | 3034 | def byProfiles(self, data, datatime): |
|
|
3035 | 3035 | |
|
|
3036 | 3036 | self.__dataReady = False |
|
|
3037 | 3037 | dataBlock = None |
|
|
3038 | 3038 | # n = None |
|
|
3039 | 3039 | # print data |
|
|
3040 | 3040 | # raise |
|
|
3041 | 3041 | self.fillBuffer(data, datatime) |
|
|
3042 | 3042 | |
|
|
3043 | 3043 | if self.__profIndex == self.n: |
|
|
3044 | 3044 | dataBlock = self.__buffer_data |
|
|
3045 | 3045 | self.__dataReady = True |
|
|
3046 | 3046 | |
|
|
3047 | 3047 | return dataBlock |
|
|
3048 | 3048 | |
|
|
3049 | 3049 | def byTime(self, data, datatime): |
|
|
3050 | 3050 | |
|
|
3051 | 3051 | self.__dataReady = False |
|
|
3052 | 3052 | dataBlock = None |
|
|
3053 | 3053 | n = None |
|
|
3054 | 3054 | |
|
|
3055 | 3055 | self.fillBuffer(data, datatime) |
|
|
3056 | 3056 | |
|
|
3057 | 3057 | if (datatime - self.__initime) >= self.__integrationtime: |
|
|
3058 | 3058 | dataBlock = self.__buffer_data |
|
|
3059 | 3059 | self.n = self.__profIndex |
|
|
3060 | 3060 | self.__dataReady = True |
|
|
3061 | 3061 | |
|
|
3062 | 3062 | return dataBlock |
|
|
3063 | 3063 | |
|
|
3064 | 3064 | |
|
|
3065 | 3065 | def run(self, dataOut, n=None, timeInterval=None, **kwargs): |
|
|
3066 | 3066 | |
|
|
3067 | 3067 | if not self.isConfig: |
|
|
3068 | 3068 | self.setup(n=n, timeInterval=timeInterval, **kwargs) |
|
|
3069 | 3069 | self.__nChannels = dataOut.nChannels |
|
|
3070 | 3070 | self.__nHeis = len(dataOut.heightList) |
|
|
3071 | 3071 | self.isConfig = True |
|
|
3072 | 3072 | |
|
|
3073 | 3073 | if dataOut.flagDataAsBlock: |
|
|
3074 | 3074 | """ |
|
|
3075 | 3075 | Si la data es leida por bloques, dimension = [nChannels, nProfiles, nHeis] |
|
|
3076 | 3076 | """ |
|
|
3077 | 3077 | raise ValueError("The data is already a block") |
|
|
3078 | 3078 | return |
|
|
3079 | 3079 | else: |
|
|
3080 | 3080 | |
|
|
3081 | 3081 | dataBlock, timeBlock = self.getData(dataOut.data, dataOut.utctime) |
|
|
3082 | 3082 | |
|
|
3083 | 3083 | |
|
|
3084 | 3084 | # print(dataOut.data.shape) |
|
|
3085 | 3085 | # dataOut.timeInterval *= n |
|
|
3086 | 3086 | dataOut.flagNoData = True |
|
|
3087 | 3087 | |
|
|
3088 | 3088 | if self.__dataReady: |
|
|
3089 | 3089 | dataOut.data = dataBlock |
|
|
3090 | 3090 | dataOut.flagDataAsBlock = True |
|
|
3091 | 3091 | dataOut.utctime = timeBlock[-1] |
|
|
3092 | 3092 | dataOut.nProfiles = self.__profIndex |
|
|
3093 | 3093 | # print avgdata, avgdatatime |
|
|
3094 | 3094 | # raise |
|
|
3095 | 3095 | dataOut.flagNoData = False |
|
|
3096 | 3096 | self.__profIndex = 0 |
|
|
3097 | 3097 | self.__initime = None |
|
|
3098 | 3098 | #update Processing Header: |
|
|
3099 | 3099 | # print(dataOut.data.shape) |
|
|
3100 | 3100 | |
|
|
3101 | 3101 | return dataOut |
|
|
3102 | 3102 | |
|
|
3103 | 3103 | |
|
|
3104 | 3104 | class remFaradayProfiles(Operation): |
|
|
3105 | 3105 | |
|
|
3106 | 3106 | def __init__(self, **kwargs): |
|
|
3107 | 3107 | |
|
|
3108 | 3108 | |
|
|
3109 | 3109 | Operation.__init__(self, **kwargs) |
|
|
3110 | 3110 | |
|
|
3111 | 3111 | self.isConfig = False |
|
|
3112 | 3112 | |
|
|
3113 | 3113 | self.nprofile2 = 0 |
|
|
3114 | 3114 | self.profile = 0 |
|
|
3115 | 3115 | self.flagRun = False |
|
|
3116 | 3116 | self.flagRemove = False |
|
|
3117 | 3117 | self.k = 0 |
|
|
3118 | 3118 | |
|
|
3119 |
def setup(self, channel,nChannels=5, nProfiles=300,nBlocks=100, nIpp2= |
|
|
|
3119 | def setup(self, channel,nChannels=5, nProfiles=300,nBlocks=100, nIpp2=300, nTx2=132, nTaus=22, offTaus=14, iTaus=8, | |
|
|
3120 | 3120 | nfft=1): |
|
|
3121 | 3121 | ''' |
|
|
3122 | 3122 | nProfiles = amisr profiles per block -> raw data |
|
|
3123 | 3123 | nIpp1 = number of profiles in one AMISR sync |
|
|
3124 | 3124 | nIpp2 = number of profiles in one Jicamarca sync |
|
|
3125 | 3125 | nTx2 = number of profiles transmited for Faraday Experiment |
|
|
3126 | 3126 | nTaus = Total profiles for lags |
|
|
3127 | 3127 | offTaus = where starts the interference, (profile) |
|
|
3128 | 3128 | iTaus = lenght of the interference |
|
|
3129 | 3129 | irepeat = number of repetition of the Taus |
|
|
3130 | 3130 | ''' |
|
|
3131 | 3131 | self.nIpp2 = nIpp2 |
|
|
3132 | 3132 | self.channel = channel |
|
|
3133 | 3133 | self.nChannels = nChannels |
|
|
3134 | 3134 | self.nTx2 = nTx2 |
|
|
3135 | 3135 | self.nTaus = nTaus |
|
|
3136 | 3136 | |
|
|
3137 | 3137 | |
|
|
3138 | 3138 | booldataset = numpy.ones( (nBlocks, nProfiles) ) |
|
|
3139 | 3139 | self.profilesFlag = None |
|
|
3140 | 3140 | #marking the afected profiles |
|
|
3141 | 3141 | f_iTaus=False |
|
|
3142 | 3142 | f_ntx = False |
|
|
3143 | 3143 | fi = 0 |
|
|
3144 | 3144 | k = 0 |
|
|
3145 | 3145 | kt =0 |
|
|
3146 | 3146 | fi_reps = 0 |
|
|
3147 | 3147 | for i in range(nBlocks): |
|
|
3148 | 3148 | for j in range(nProfiles): |
|
|
3149 | 3149 | # fi 0---nTaus |
|
|
3150 | 3150 | # |
|
|
3151 | 3151 | if k%nIpp2==0: #each sync PPs or 2, 3, or 5 |
|
|
3152 | 3152 | f_ntx = True |
|
|
3153 | 3153 | kt = 0 |
|
|
3154 | #print(k, fi, j, f_iTaus) | |
|
|
3155 | 3154 | if f_ntx: |
|
|
3156 | 3155 | |
|
|
3157 | 3156 | if kt%nTaus==0: #each sequence of Taus |
|
|
3158 | 3157 | f_iTaus = True |
|
|
3159 | 3158 | fi = 0 |
|
|
3160 | 3159 | |
|
|
3161 | 3160 | if f_iTaus: |
|
|
3162 | 3161 | if fi > offTaus-1: |
|
|
3163 | 3162 | booldataset[i, j]=0 #Afected profile |
|
|
3164 | 3163 | fi += 1 |
|
|
3165 | 3164 | if fi == nTaus-1: #restart the taus sequence |
|
|
3166 | 3165 | fi = 0 |
|
|
3167 | 3166 | f_iTaus = False |
|
|
3168 | 3167 | fi_reps += 1 |
|
|
3169 | 3168 | if fi_reps == (nTx2/nTaus): |
|
|
3170 | 3169 | fi = 0 |
|
|
3171 | #print("AQUI, ", fi_reps, k, fi) | |
|
|
3172 | 3170 | fi_reps = 0 |
|
|
3173 | 3171 | f_ntx=False |
|
|
3174 | # if i < 1: | |
|
|
3175 | # print(fi, kt) | |
|
|
3176 | 3172 | kt += 1 |
|
|
3177 | 3173 | k += 1 |
|
|
3178 | 3174 | |
|
|
3179 | 3175 | # fig = plt.figure() |
|
|
3180 | 3176 | # ax = fig.add_subplot(111) |
|
|
3181 | 3177 | # cax = ax.pcolormesh(booldataset, cmap='plasma') |
|
|
3182 | 3178 | # cbar = fig.colorbar(cax) |
|
|
3183 | 3179 | # plt.show() |
|
|
3184 | 3180 | |
|
|
3185 | ||
|
|
3186 | #print ("AQUI") | |
|
|
3181 | ||
|
|
3187 | 3182 | #reshape the Flag as AMISR reader |
|
|
3188 | 3183 | |
|
|
3189 | 3184 | profPerCH = int( (nProfiles) / (nfft*nChannels)) |
|
|
3190 | 3185 | new_block = numpy.empty( (nBlocks, nChannels, int(nProfiles/nChannels) ) ) |
|
|
3191 | 3186 | # print(new_block.shape, profPerCH) |
|
|
3192 | 3187 | for thisChannel in range(nChannels): |
|
|
3193 | 3188 | |
|
|
3194 | 3189 | ich = thisChannel |
|
|
3195 | 3190 | |
|
|
3196 | 3191 | idx_ch = [nfft*(ich + nChannels*k) for k in range(profPerCH)] |
|
|
3197 | 3192 | #print(idx_ch) |
|
|
3198 | 3193 | if nfft > 1: |
|
|
3199 | 3194 | aux = [numpy.arange(i, i+nfft) for i in idx_ch] |
|
|
3200 | 3195 | idx_ch = None |
|
|
3201 | 3196 | idx_ch =aux |
|
|
3202 | 3197 | idx_ch = numpy.array(idx_ch, dtype=int).flatten() |
|
|
3203 | 3198 | else: |
|
|
3204 | 3199 | idx_ch = numpy.array(idx_ch, dtype=int) |
|
|
3205 | 3200 | |
|
|
3206 | 3201 | new_block[:,ich,:] = booldataset[:,idx_ch] |
|
|
3207 | 3202 | |
|
|
3208 | 3203 | new_block = numpy.transpose(new_block, (1,0,2)) |
|
|
3209 | new_block = numpy.reshape(new_block, (nChannels,-1)) | |
|
|
3210 |
|
|
|
|
3204 | #new_block = numpy.reshape(new_block, (nChannels,-1)) | |
|
|
3205 | new_block = numpy.reshape(new_block, (nChannels,profPerCH*nBlocks)) | |
|
|
3211 | 3206 | self.profilesFlag = new_block.copy() |
|
|
3212 | 3207 | |
|
|
3213 | 3208 | # fig = plt.figure() |
|
|
3214 | 3209 | # ax = fig.add_subplot(111) |
|
|
3215 | 3210 | # cax = ax.pcolormesh(new_block, cmap='plasma') |
|
|
3216 | 3211 | # cbar = fig.colorbar(cax) |
|
|
3217 | 3212 | # plt.show() |
|
|
3218 | 3213 | |
|
|
3219 | 3214 | self.isConfig = True |
|
|
3220 | 3215 | |
|
|
3221 | #print(self.profilesFlag.shape) | |
|
|
3222 | 3216 | |
|
|
3223 | 3217 | def run(self,dataOut, channel=0, nChannels=5, nProfiles=300,nBlocks=100,nIpp1=100, |
|
|
3224 | 3218 | nIpp2=300, nTx2=132, nTaus=22, offTaus=8, iTaus=14, nfft=1 ,offIpp=0): |
|
|
3225 | 3219 | |
|
|
3226 | 3220 | dataOut.flagNoData = False |
|
|
3227 | 3221 | |
|
|
3228 | 3222 | if not self.isConfig: |
|
|
3229 | 3223 | self.setup(channel,nChannels=nChannels, nProfiles=nProfiles,nBlocks=nBlocks, nIpp2=nIpp2, |
|
|
3230 | 3224 | nTx2=nTx2, nTaus=nTaus, offTaus=offTaus, iTaus=iTaus, nfft=nfft) |
|
|
3231 | 3225 | #print("Setup Done") |
|
|
3232 | 3226 | #print(offIpp*nIpp1/nChannels) |
|
|
3233 | 3227 | if not self.flagRun: |
|
|
3234 | 3228 | if self.nprofile2 < offIpp*nIpp1/nChannels : |
|
|
3235 | 3229 | self.nprofile2 += 1 |
|
|
3236 | 3230 | return dataOut |
|
|
3237 | 3231 | else: |
|
|
3238 | 3232 | self.flagRun = True |
|
|
3239 | 3233 | self.profile = 0 |
|
|
3240 | ||
|
|
3241 | 3234 | |
|
|
3242 | 3235 | #check profile ## Faraday interference |
|
|
3243 | 3236 | if self.profilesFlag[channel, self.profile]==0: |
|
|
3244 | 3237 | dataOut.flagNoData = True # do not pass this profile |
|
|
3245 | # print(self.nprofile, dataOut.flagNoData) | |
|
|
3246 | #print(self.nprofile2, self.profile, dataOut.flagNoData) | |
|
|
3238 | ||
|
|
3247 | 3239 | self.profile +=1 |
|
|
3248 | # if self.profile == int((nProfiles*nBlocks)/self.nChannels): | |
|
|
3249 | # self.flagRun=False | |
|
|
3250 | # self.profile = 0 | |
|
|
3240 | ||
|
|
3251 | 3241 | |
|
|
3252 | 3242 | self.nprofile2 +=1 |
|
|
3253 | 3243 | |
|
|
3254 | 3244 | if self.nprofile2 == int((nProfiles*nBlocks)/self.nChannels): |
|
|
3255 | 3245 | self.nprofile2 = 0 |
|
|
3256 | 3246 | self.profile = 0 |
|
|
3257 | 3247 | self.flagRun = False |
|
|
3258 | 3248 | |
|
|
3259 | 3249 | return dataOut No newline at end of file |
General Comments 0
You need to be logged in to leave comments.
Login now