##// END OF EJS Templates
schain
RSS
minor changes
Miguel Valdez -
r750:e82e20d80835
parent child
Show More
Show file before | Show file after | Hide comments
@@ -1,904 +1,904
1 1 import numpy
2 2
3 3 from jroproc_base import ProcessingUnit, Operation
4 4 from schainpy.model.data.jrodata import Spectra
5 5 from schainpy.model.data.jrodata import hildebrand_sekhon
6 6
7 7 class SpectraProc(ProcessingUnit):
8 8
9 9 def __init__(self):
10 10
11 11 ProcessingUnit.__init__(self)
12 12
13 13 self.buffer = None
14 14 self.firstdatatime = None
15 15 self.profIndex = 0
16 16 self.dataOut = Spectra()
17 17 self.id_min = None
18 18 self.id_max = None
19 19
20 20 def __updateSpecFromVoltage(self):
21 21
22 22 self.dataOut.timeZone = self.dataIn.timeZone
23 23 self.dataOut.dstFlag = self.dataIn.dstFlag
24 24 self.dataOut.errorCount = self.dataIn.errorCount
25 25 self.dataOut.useLocalTime = self.dataIn.useLocalTime
26 26
27 27 self.dataOut.radarControllerHeaderObj = self.dataIn.radarControllerHeaderObj.copy()
28 28 self.dataOut.systemHeaderObj = self.dataIn.systemHeaderObj.copy()
29 29 self.dataOut.channelList = self.dataIn.channelList
30 30 self.dataOut.heightList = self.dataIn.heightList
31 31 self.dataOut.dtype = numpy.dtype([('real','<f4'),('imag','<f4')])
32 32
33 33 self.dataOut.nBaud = self.dataIn.nBaud
34 34 self.dataOut.nCode = self.dataIn.nCode
35 35 self.dataOut.code = self.dataIn.code
36 36 self.dataOut.nProfiles = self.dataOut.nFFTPoints
37 37
38 38 self.dataOut.flagDiscontinuousBlock = self.dataIn.flagDiscontinuousBlock
39 39 self.dataOut.utctime = self.firstdatatime
40 40 self.dataOut.flagDecodeData = self.dataIn.flagDecodeData #asumo q la data esta decodificada
41 41 self.dataOut.flagDeflipData = self.dataIn.flagDeflipData #asumo q la data esta sin flip
42 42 self.dataOut.flagShiftFFT = False
43 43
44 44 self.dataOut.nCohInt = self.dataIn.nCohInt
45 45 self.dataOut.nIncohInt = 1
46 46
47 47 self.dataOut.windowOfFilter = self.dataIn.windowOfFilter
48 48
49 49 self.dataOut.frequency = self.dataIn.frequency
50 50 self.dataOut.realtime = self.dataIn.realtime
51 51
52 52 self.dataOut.azimuth = self.dataIn.azimuth
53 53 self.dataOut.zenith = self.dataIn.zenith
54 54
55 55 self.dataOut.beam.codeList = self.dataIn.beam.codeList
56 56 self.dataOut.beam.azimuthList = self.dataIn.beam.azimuthList
57 57 self.dataOut.beam.zenithList = self.dataIn.beam.zenithList
58 58
59 59 def __getFft(self):
60 60 """
61 61 Convierte valores de Voltaje a Spectra
62 62
63 63 Affected:
64 64 self.dataOut.data_spc
65 65 self.dataOut.data_cspc
66 66 self.dataOut.data_dc
67 67 self.dataOut.heightList
68 68 self.profIndex
69 69 self.buffer
70 70 self.dataOut.flagNoData
71 71 """
72 72 fft_volt = numpy.fft.fft(self.buffer,n=self.dataOut.nFFTPoints,axis=1)
73 73 fft_volt = fft_volt.astype(numpy.dtype('complex'))
74 74 dc = fft_volt[:,0,:]
75 75
76 76 #calculo de self-spectra
77 77 fft_volt = numpy.fft.fftshift(fft_volt,axes=(1,))
78 78 spc = fft_volt * numpy.conjugate(fft_volt)
79 79 spc = spc.real
80 80
81 81 blocksize = 0
82 82 blocksize += dc.size
83 83 blocksize += spc.size
84 84
85 85 cspc = None
86 86 pairIndex = 0
87 87 if self.dataOut.pairsList != None:
88 88 #calculo de cross-spectra
89 89 cspc = numpy.zeros((self.dataOut.nPairs, self.dataOut.nFFTPoints, self.dataOut.nHeights), dtype='complex')
90 90 for pair in self.dataOut.pairsList:
91 91 if pair[0] not in self.dataOut.channelList:
92 92 raise ValueError, "Error getting CrossSpectra: pair 0 of %s is not in channelList = %s" %(str(pair), str(self.dataOut.channelList))
93 93 if pair[1] not in self.dataOut.channelList:
94 94 raise ValueError, "Error getting CrossSpectra: pair 1 of %s is not in channelList = %s" %(str(pair), str(self.dataOut.channelList))
95 95
96 96 cspc[pairIndex,:,:] = fft_volt[pair[0],:,:] * numpy.conjugate(fft_volt[pair[1],:,:])
97 97 pairIndex += 1
98 98 blocksize += cspc.size
99 99
100 100 self.dataOut.data_spc = spc
101 101 self.dataOut.data_cspc = cspc
102 102 self.dataOut.data_dc = dc
103 103 self.dataOut.blockSize = blocksize
104 104 self.dataOut.flagShiftFFT = True
105 105
106 106 def run(self, nProfiles=None, nFFTPoints=None, pairsList=[], ippFactor=None):
107 107
108 108 self.dataOut.flagNoData = True
109 109
110 110 if self.dataIn.type == "Spectra":
111 111 self.dataOut.copy(self.dataIn)
112 112 # self.__selectPairs(pairsList)
113 113 return True
114 114
115 115 if self.dataIn.type == "Voltage":
116 116
117 117 if nFFTPoints == None:
118 118 raise ValueError, "This SpectraProc.run() need nFFTPoints input variable"
119 119
120 120 if nProfiles == None:
121 121 nProfiles = nFFTPoints
122 122
123 123 if ippFactor == None:
124 124 ippFactor = 1
125 125
126 126 self.dataOut.ippFactor = ippFactor
127 127
128 128 self.dataOut.nFFTPoints = nFFTPoints
129 129 self.dataOut.pairsList = pairsList
130 130
131 131 if self.buffer is None:
132 132 self.buffer = numpy.zeros( (self.dataIn.nChannels,
133 133 nProfiles,
134 134 self.dataIn.nHeights),
135 135 dtype='complex')
136 136
137 137 if self.dataIn.flagDataAsBlock:
138 138 #data dimension: [nChannels, nProfiles, nSamples]
139 139 nVoltProfiles = self.dataIn.data.shape[1]
140 140 # nVoltProfiles = self.dataIn.nProfiles
141 141
142 142 if nVoltProfiles == nProfiles:
143 143 self.buffer = self.dataIn.data.copy()
144 144 self.profIndex = nVoltProfiles
145 145
146 146 elif nVoltProfiles < nProfiles:
147 147
148 148 if self.profIndex == 0:
149 149 self.id_min = 0
150 150 self.id_max = nVoltProfiles
151 151
152 152 self.buffer[:,self.id_min:self.id_max,:] = self.dataIn.data
153 153 self.profIndex += nVoltProfiles
154 154 self.id_min += nVoltProfiles
155 155 self.id_max += nVoltProfiles
156 156 else:
157 raise ValueError, "The type object %s has %d profiles, it should be equal to %d profiles"%(self.dataIn.type,self.dataIn.data.shape[1],nProfiles)
157 raise ValueError, "The type object %s has %d profiles, it should just has %d profiles"%(self.dataIn.type,self.dataIn.data.shape[1],nProfiles)
158 158 self.dataOut.flagNoData = True
159 159 return 0
160 160 else:
161 161 self.buffer[:,self.profIndex,:] = self.dataIn.data.copy()
162 162 self.profIndex += 1
163 163
164 164 if self.firstdatatime == None:
165 165 self.firstdatatime = self.dataIn.utctime
166 166
167 167 if self.profIndex == nProfiles:
168 168 self.__updateSpecFromVoltage()
169 169 self.__getFft()
170 170
171 171 self.dataOut.flagNoData = False
172 172 self.firstdatatime = None
173 173 self.profIndex = 0
174 174
175 175 return True
176 176
177 177 raise ValueError, "The type of input object '%s' is not valid"%(self.dataIn.type)
178 178
179 179 def __selectPairs(self, pairsList):
180 180
181 181 if channelList == None:
182 182 return
183 183
184 184 pairsIndexListSelected = []
185 185
186 186 for thisPair in pairsList:
187 187
188 188 if thisPair not in self.dataOut.pairsList:
189 189 continue
190 190
191 191 pairIndex = self.dataOut.pairsList.index(thisPair)
192 192
193 193 pairsIndexListSelected.append(pairIndex)
194 194
195 195 if not pairsIndexListSelected:
196 196 self.dataOut.data_cspc = None
197 197 self.dataOut.pairsList = []
198 198 return
199 199
200 200 self.dataOut.data_cspc = self.dataOut.data_cspc[pairsIndexListSelected]
201 201 self.dataOut.pairsList = [self.dataOut.pairsList[i] for i in pairsIndexListSelected]
202 202
203 203 return
204 204
205 205 def __selectPairsByChannel(self, channelList=None):
206 206
207 207 if channelList == None:
208 208 return
209 209
210 210 pairsIndexListSelected = []
211 211 for pairIndex in self.dataOut.pairsIndexList:
212 212 #First pair
213 213 if self.dataOut.pairsList[pairIndex][0] not in channelList:
214 214 continue
215 215 #Second pair
216 216 if self.dataOut.pairsList[pairIndex][1] not in channelList:
217 217 continue
218 218
219 219 pairsIndexListSelected.append(pairIndex)
220 220
221 221 if not pairsIndexListSelected:
222 222 self.dataOut.data_cspc = None
223 223 self.dataOut.pairsList = []
224 224 return
225 225
226 226 self.dataOut.data_cspc = self.dataOut.data_cspc[pairsIndexListSelected]
227 227 self.dataOut.pairsList = [self.dataOut.pairsList[i] for i in pairsIndexListSelected]
228 228
229 229 return
230 230
231 231 def selectChannels(self, channelList):
232 232
233 233 channelIndexList = []
234 234
235 235 for channel in channelList:
236 236 if channel not in self.dataOut.channelList:
237 237 raise ValueError, "Error selecting channels, Channel %d is not valid.\nAvailable channels = %s" %(channel, str(self.dataOut.channelList))
238 238
239 239 index = self.dataOut.channelList.index(channel)
240 240 channelIndexList.append(index)
241 241
242 242 self.selectChannelsByIndex(channelIndexList)
243 243
244 244 def selectChannelsByIndex(self, channelIndexList):
245 245 """
246 246 Selecciona un bloque de datos en base a canales segun el channelIndexList
247 247
248 248 Input:
249 249 channelIndexList : lista sencilla de canales a seleccionar por ej. [2,3,7]
250 250
251 251 Affected:
252 252 self.dataOut.data_spc
253 253 self.dataOut.channelIndexList
254 254 self.dataOut.nChannels
255 255
256 256 Return:
257 257 None
258 258 """
259 259
260 260 for channelIndex in channelIndexList:
261 261 if channelIndex not in self.dataOut.channelIndexList:
262 262 raise ValueError, "Error selecting channels: The value %d in channelIndexList is not valid.\nAvailable channel indexes = " %(channelIndex, self.dataOut.channelIndexList)
263 263
264 264 # nChannels = len(channelIndexList)
265 265
266 266 data_spc = self.dataOut.data_spc[channelIndexList,:]
267 267 data_dc = self.dataOut.data_dc[channelIndexList,:]
268 268
269 269 self.dataOut.data_spc = data_spc
270 270 self.dataOut.data_dc = data_dc
271 271
272 272 self.dataOut.channelList = [self.dataOut.channelList[i] for i in channelIndexList]
273 273 # self.dataOut.nChannels = nChannels
274 274
275 275 self.__selectPairsByChannel(self.dataOut.channelList)
276 276
277 277 return 1
278 278
279 279 def selectHeights(self, minHei, maxHei):
280 280 """
281 281 Selecciona un bloque de datos en base a un grupo de valores de alturas segun el rango
282 282 minHei <= height <= maxHei
283 283
284 284 Input:
285 285 minHei : valor minimo de altura a considerar
286 286 maxHei : valor maximo de altura a considerar
287 287
288 288 Affected:
289 289 Indirectamente son cambiados varios valores a travez del metodo selectHeightsByIndex
290 290
291 291 Return:
292 292 1 si el metodo se ejecuto con exito caso contrario devuelve 0
293 293 """
294 294
295 295 if (minHei > maxHei):
296 296 raise ValueError, "Error selecting heights: Height range (%d,%d) is not valid" % (minHei, maxHei)
297 297
298 298 if (minHei < self.dataOut.heightList[0]):
299 299 minHei = self.dataOut.heightList[0]
300 300
301 301 if (maxHei > self.dataOut.heightList[-1]):
302 302 maxHei = self.dataOut.heightList[-1]
303 303
304 304 minIndex = 0
305 305 maxIndex = 0
306 306 heights = self.dataOut.heightList
307 307
308 308 inda = numpy.where(heights >= minHei)
309 309 indb = numpy.where(heights <= maxHei)
310 310
311 311 try:
312 312 minIndex = inda[0][0]
313 313 except:
314 314 minIndex = 0
315 315
316 316 try:
317 317 maxIndex = indb[0][-1]
318 318 except:
319 319 maxIndex = len(heights)
320 320
321 321 self.selectHeightsByIndex(minIndex, maxIndex)
322 322
323 323 return 1
324 324
325 325 def getBeaconSignal(self, tauindex = 0, channelindex = 0, hei_ref=None):
326 326 newheis = numpy.where(self.dataOut.heightList>self.dataOut.radarControllerHeaderObj.Taus[tauindex])
327 327
328 328 if hei_ref != None:
329 329 newheis = numpy.where(self.dataOut.heightList>hei_ref)
330 330
331 331 minIndex = min(newheis[0])
332 332 maxIndex = max(newheis[0])
333 333 data_spc = self.dataOut.data_spc[:,:,minIndex:maxIndex+1]
334 334 heightList = self.dataOut.heightList[minIndex:maxIndex+1]
335 335
336 336 # determina indices
337 337 nheis = int(self.dataOut.radarControllerHeaderObj.txB/(self.dataOut.heightList[1]-self.dataOut.heightList[0]))
338 338 avg_dB = 10*numpy.log10(numpy.sum(data_spc[channelindex,:,:],axis=0))
339 339 beacon_dB = numpy.sort(avg_dB)[-nheis:]
340 340 beacon_heiIndexList = []
341 341 for val in avg_dB.tolist():
342 342 if val >= beacon_dB[0]:
343 343 beacon_heiIndexList.append(avg_dB.tolist().index(val))
344 344
345 345 #data_spc = data_spc[:,:,beacon_heiIndexList]
346 346 data_cspc = None
347 347 if self.dataOut.data_cspc is not None:
348 348 data_cspc = self.dataOut.data_cspc[:,:,minIndex:maxIndex+1]
349 349 #data_cspc = data_cspc[:,:,beacon_heiIndexList]
350 350
351 351 data_dc = None
352 352 if self.dataOut.data_dc is not None:
353 353 data_dc = self.dataOut.data_dc[:,minIndex:maxIndex+1]
354 354 #data_dc = data_dc[:,beacon_heiIndexList]
355 355
356 356 self.dataOut.data_spc = data_spc
357 357 self.dataOut.data_cspc = data_cspc
358 358 self.dataOut.data_dc = data_dc
359 359 self.dataOut.heightList = heightList
360 360 self.dataOut.beacon_heiIndexList = beacon_heiIndexList
361 361
362 362 return 1
363 363
364 364
365 365 def selectHeightsByIndex(self, minIndex, maxIndex):
366 366 """
367 367 Selecciona un bloque de datos en base a un grupo indices de alturas segun el rango
368 368 minIndex <= index <= maxIndex
369 369
370 370 Input:
371 371 minIndex : valor de indice minimo de altura a considerar
372 372 maxIndex : valor de indice maximo de altura a considerar
373 373
374 374 Affected:
375 375 self.dataOut.data_spc
376 376 self.dataOut.data_cspc
377 377 self.dataOut.data_dc
378 378 self.dataOut.heightList
379 379
380 380 Return:
381 381 1 si el metodo se ejecuto con exito caso contrario devuelve 0
382 382 """
383 383
384 384 if (minIndex < 0) or (minIndex > maxIndex):
385 385 raise ValueError, "Error selecting heights: Index range (%d,%d) is not valid" % (minIndex, maxIndex)
386 386
387 387 if (maxIndex >= self.dataOut.nHeights):
388 388 maxIndex = self.dataOut.nHeights-1
389 389
390 390 #Spectra
391 391 data_spc = self.dataOut.data_spc[:,:,minIndex:maxIndex+1]
392 392
393 393 data_cspc = None
394 394 if self.dataOut.data_cspc is not None:
395 395 data_cspc = self.dataOut.data_cspc[:,:,minIndex:maxIndex+1]
396 396
397 397 data_dc = None
398 398 if self.dataOut.data_dc is not None:
399 399 data_dc = self.dataOut.data_dc[:,minIndex:maxIndex+1]
400 400
401 401 self.dataOut.data_spc = data_spc
402 402 self.dataOut.data_cspc = data_cspc
403 403 self.dataOut.data_dc = data_dc
404 404
405 405 self.dataOut.heightList = self.dataOut.heightList[minIndex:maxIndex+1]
406 406
407 407 return 1
408 408
409 409 def removeDC(self, mode = 2):
410 410 jspectra = self.dataOut.data_spc
411 411 jcspectra = self.dataOut.data_cspc
412 412
413 413
414 414 num_chan = jspectra.shape[0]
415 415 num_hei = jspectra.shape[2]
416 416
417 417 if jcspectra is not None:
418 418 jcspectraExist = True
419 419 num_pairs = jcspectra.shape[0]
420 420 else: jcspectraExist = False
421 421
422 422 freq_dc = jspectra.shape[1]/2
423 423 ind_vel = numpy.array([-2,-1,1,2]) + freq_dc
424 424
425 425 if ind_vel[0]<0:
426 426 ind_vel[range(0,1)] = ind_vel[range(0,1)] + self.num_prof
427 427
428 428 if mode == 1:
429 429 jspectra[:,freq_dc,:] = (jspectra[:,ind_vel[1],:] + jspectra[:,ind_vel[2],:])/2 #CORRECCION
430 430
431 431 if jcspectraExist:
432 432 jcspectra[:,freq_dc,:] = (jcspectra[:,ind_vel[1],:] + jcspectra[:,ind_vel[2],:])/2
433 433
434 434 if mode == 2:
435 435
436 436 vel = numpy.array([-2,-1,1,2])
437 437 xx = numpy.zeros([4,4])
438 438
439 439 for fil in range(4):
440 440 xx[fil,:] = vel[fil]**numpy.asarray(range(4))
441 441
442 442 xx_inv = numpy.linalg.inv(xx)
443 443 xx_aux = xx_inv[0,:]
444 444
445 445 for ich in range(num_chan):
446 446 yy = jspectra[ich,ind_vel,:]
447 447 jspectra[ich,freq_dc,:] = numpy.dot(xx_aux,yy)
448 448
449 449 junkid = jspectra[ich,freq_dc,:]<=0
450 450 cjunkid = sum(junkid)
451 451
452 452 if cjunkid.any():
453 453 jspectra[ich,freq_dc,junkid.nonzero()] = (jspectra[ich,ind_vel[1],junkid] + jspectra[ich,ind_vel[2],junkid])/2
454 454
455 455 if jcspectraExist:
456 456 for ip in range(num_pairs):
457 457 yy = jcspectra[ip,ind_vel,:]
458 458 jcspectra[ip,freq_dc,:] = numpy.dot(xx_aux,yy)
459 459
460 460
461 461 self.dataOut.data_spc = jspectra
462 462 self.dataOut.data_cspc = jcspectra
463 463
464 464 return 1
465 465
466 466 def removeInterference(self, interf = 2,hei_interf = None, nhei_interf = None, offhei_interf = None):
467 467
468 468 jspectra = self.dataOut.data_spc
469 469 jcspectra = self.dataOut.data_cspc
470 470 jnoise = self.dataOut.getNoise()
471 471 num_incoh = self.dataOut.nIncohInt
472 472
473 473 num_channel = jspectra.shape[0]
474 474 num_prof = jspectra.shape[1]
475 475 num_hei = jspectra.shape[2]
476 476
477 477 #hei_interf
478 478 if hei_interf is None:
479 479 count_hei = num_hei/2 #Como es entero no importa
480 480 hei_interf = numpy.asmatrix(range(count_hei)) + num_hei - count_hei
481 481 hei_interf = numpy.asarray(hei_interf)[0]
482 482 #nhei_interf
483 483 if (nhei_interf == None):
484 484 nhei_interf = 5
485 485 if (nhei_interf < 1):
486 486 nhei_interf = 1
487 487 if (nhei_interf > count_hei):
488 488 nhei_interf = count_hei
489 489 if (offhei_interf == None):
490 490 offhei_interf = 0
491 491
492 492 ind_hei = range(num_hei)
493 493 # mask_prof = numpy.asarray(range(num_prof - 2)) + 1
494 494 # mask_prof[range(num_prof/2 - 1,len(mask_prof))] += 1
495 495 mask_prof = numpy.asarray(range(num_prof))
496 496 num_mask_prof = mask_prof.size
497 497 comp_mask_prof = [0, num_prof/2]
498 498
499 499
500 500 #noise_exist: Determina si la variable jnoise ha sido definida y contiene la informacion del ruido de cada canal
501 501 if (jnoise.size < num_channel or numpy.isnan(jnoise).any()):
502 502 jnoise = numpy.nan
503 503 noise_exist = jnoise[0] < numpy.Inf
504 504
505 505 #Subrutina de Remocion de la Interferencia
506 506 for ich in range(num_channel):
507 507 #Se ordena los espectros segun su potencia (menor a mayor)
508 508 power = jspectra[ich,mask_prof,:]
509 509 power = power[:,hei_interf]
510 510 power = power.sum(axis = 0)
511 511 psort = power.ravel().argsort()
512 512
513 513 #Se estima la interferencia promedio en los Espectros de Potencia empleando
514 514 junkspc_interf = jspectra[ich,:,hei_interf[psort[range(offhei_interf, nhei_interf + offhei_interf)]]]
515 515
516 516 if noise_exist:
517 517 # tmp_noise = jnoise[ich] / num_prof
518 518 tmp_noise = jnoise[ich]
519 519 junkspc_interf = junkspc_interf - tmp_noise
520 520 #junkspc_interf[:,comp_mask_prof] = 0
521 521
522 522 jspc_interf = junkspc_interf.sum(axis = 0) / nhei_interf
523 523 jspc_interf = jspc_interf.transpose()
524 524 #Calculando el espectro de interferencia promedio
525 525 noiseid = numpy.where(jspc_interf <= tmp_noise/ numpy.sqrt(num_incoh))
526 526 noiseid = noiseid[0]
527 527 cnoiseid = noiseid.size
528 528 interfid = numpy.where(jspc_interf > tmp_noise/ numpy.sqrt(num_incoh))
529 529 interfid = interfid[0]
530 530 cinterfid = interfid.size
531 531
532 532 if (cnoiseid > 0): jspc_interf[noiseid] = 0
533 533
534 534 #Expandiendo los perfiles a limpiar
535 535 if (cinterfid > 0):
536 536 new_interfid = (numpy.r_[interfid - 1, interfid, interfid + 1] + num_prof)%num_prof
537 537 new_interfid = numpy.asarray(new_interfid)
538 538 new_interfid = {x for x in new_interfid}
539 539 new_interfid = numpy.array(list(new_interfid))
540 540 new_cinterfid = new_interfid.size
541 541 else: new_cinterfid = 0
542 542
543 543 for ip in range(new_cinterfid):
544 544 ind = junkspc_interf[:,new_interfid[ip]].ravel().argsort()
545 545 jspc_interf[new_interfid[ip]] = junkspc_interf[ind[nhei_interf/2],new_interfid[ip]]
546 546
547 547
548 548 jspectra[ich,:,ind_hei] = jspectra[ich,:,ind_hei] - jspc_interf #Corregir indices
549 549
550 550 #Removiendo la interferencia del punto de mayor interferencia
551 551 ListAux = jspc_interf[mask_prof].tolist()
552 552 maxid = ListAux.index(max(ListAux))
553 553
554 554
555 555 if cinterfid > 0:
556 556 for ip in range(cinterfid*(interf == 2) - 1):
557 557 ind = (jspectra[ich,interfid[ip],:] < tmp_noise*(1 + 1/numpy.sqrt(num_incoh))).nonzero()
558 558 cind = len(ind)
559 559
560 560 if (cind > 0):
561 561 jspectra[ich,interfid[ip],ind] = tmp_noise*(1 + (numpy.random.uniform(cind) - 0.5)/numpy.sqrt(num_incoh))
562 562
563 563 ind = numpy.array([-2,-1,1,2])
564 564 xx = numpy.zeros([4,4])
565 565
566 566 for id1 in range(4):
567 567 xx[:,id1] = ind[id1]**numpy.asarray(range(4))
568 568
569 569 xx_inv = numpy.linalg.inv(xx)
570 570 xx = xx_inv[:,0]
571 571 ind = (ind + maxid + num_mask_prof)%num_mask_prof
572 572 yy = jspectra[ich,mask_prof[ind],:]
573 573 jspectra[ich,mask_prof[maxid],:] = numpy.dot(yy.transpose(),xx)
574 574
575 575
576 576 indAux = (jspectra[ich,:,:] < tmp_noise*(1-1/numpy.sqrt(num_incoh))).nonzero()
577 577 jspectra[ich,indAux[0],indAux[1]] = tmp_noise * (1 - 1/numpy.sqrt(num_incoh))
578 578
579 579 #Remocion de Interferencia en el Cross Spectra
580 580 if jcspectra is None: return jspectra, jcspectra
581 581 num_pairs = jcspectra.size/(num_prof*num_hei)
582 582 jcspectra = jcspectra.reshape(num_pairs, num_prof, num_hei)
583 583
584 584 for ip in range(num_pairs):
585 585
586 586 #-------------------------------------------
587 587
588 588 cspower = numpy.abs(jcspectra[ip,mask_prof,:])
589 589 cspower = cspower[:,hei_interf]
590 590 cspower = cspower.sum(axis = 0)
591 591
592 592 cspsort = cspower.ravel().argsort()
593 593 junkcspc_interf = jcspectra[ip,:,hei_interf[cspsort[range(offhei_interf, nhei_interf + offhei_interf)]]]
594 594 junkcspc_interf = junkcspc_interf.transpose()
595 595 jcspc_interf = junkcspc_interf.sum(axis = 1)/nhei_interf
596 596
597 597 ind = numpy.abs(jcspc_interf[mask_prof]).ravel().argsort()
598 598
599 599 median_real = numpy.median(numpy.real(junkcspc_interf[mask_prof[ind[range(3*num_prof/4)]],:]))
600 600 median_imag = numpy.median(numpy.imag(junkcspc_interf[mask_prof[ind[range(3*num_prof/4)]],:]))
601 601 junkcspc_interf[comp_mask_prof,:] = numpy.complex(median_real, median_imag)
602 602
603 603 for iprof in range(num_prof):
604 604 ind = numpy.abs(junkcspc_interf[iprof,:]).ravel().argsort()
605 605 jcspc_interf[iprof] = junkcspc_interf[iprof, ind[nhei_interf/2]]
606 606
607 607 #Removiendo la Interferencia
608 608 jcspectra[ip,:,ind_hei] = jcspectra[ip,:,ind_hei] - jcspc_interf
609 609
610 610 ListAux = numpy.abs(jcspc_interf[mask_prof]).tolist()
611 611 maxid = ListAux.index(max(ListAux))
612 612
613 613 ind = numpy.array([-2,-1,1,2])
614 614 xx = numpy.zeros([4,4])
615 615
616 616 for id1 in range(4):
617 617 xx[:,id1] = ind[id1]**numpy.asarray(range(4))
618 618
619 619 xx_inv = numpy.linalg.inv(xx)
620 620 xx = xx_inv[:,0]
621 621
622 622 ind = (ind + maxid + num_mask_prof)%num_mask_prof
623 623 yy = jcspectra[ip,mask_prof[ind],:]
624 624 jcspectra[ip,mask_prof[maxid],:] = numpy.dot(yy.transpose(),xx)
625 625
626 626 #Guardar Resultados
627 627 self.dataOut.data_spc = jspectra
628 628 self.dataOut.data_cspc = jcspectra
629 629
630 630 return 1
631 631
632 632 def setRadarFrequency(self, frequency=None):
633 633
634 634 if frequency != None:
635 635 self.dataOut.frequency = frequency
636 636
637 637 return 1
638 638
639 639 def getNoise(self, minHei=None, maxHei=None, minVel=None, maxVel=None):
640 640 #validacion de rango
641 641 if minHei == None:
642 642 minHei = self.dataOut.heightList[0]
643 643
644 644 if maxHei == None:
645 645 maxHei = self.dataOut.heightList[-1]
646 646
647 647 if (minHei < self.dataOut.heightList[0]) or (minHei > maxHei):
648 648 print 'minHei: %.2f is out of the heights range'%(minHei)
649 649 print 'minHei is setting to %.2f'%(self.dataOut.heightList[0])
650 650 minHei = self.dataOut.heightList[0]
651 651
652 652 if (maxHei > self.dataOut.heightList[-1]) or (maxHei < minHei):
653 653 print 'maxHei: %.2f is out of the heights range'%(maxHei)
654 654 print 'maxHei is setting to %.2f'%(self.dataOut.heightList[-1])
655 655 maxHei = self.dataOut.heightList[-1]
656 656
657 657 # validacion de velocidades
658 658 velrange = self.dataOut.getVelRange(1)
659 659
660 660 if minVel == None:
661 661 minVel = velrange[0]
662 662
663 663 if maxVel == None:
664 664 maxVel = velrange[-1]
665 665
666 666 if (minVel < velrange[0]) or (minVel > maxVel):
667 667 print 'minVel: %.2f is out of the velocity range'%(minVel)
668 668 print 'minVel is setting to %.2f'%(velrange[0])
669 669 minVel = velrange[0]
670 670
671 671 if (maxVel > velrange[-1]) or (maxVel < minVel):
672 672 print 'maxVel: %.2f is out of the velocity range'%(maxVel)
673 673 print 'maxVel is setting to %.2f'%(velrange[-1])
674 674 maxVel = velrange[-1]
675 675
676 676 # seleccion de indices para rango
677 677 minIndex = 0
678 678 maxIndex = 0
679 679 heights = self.dataOut.heightList
680 680
681 681 inda = numpy.where(heights >= minHei)
682 682 indb = numpy.where(heights <= maxHei)
683 683
684 684 try:
685 685 minIndex = inda[0][0]
686 686 except:
687 687 minIndex = 0
688 688
689 689 try:
690 690 maxIndex = indb[0][-1]
691 691 except:
692 692 maxIndex = len(heights)
693 693
694 694 if (minIndex < 0) or (minIndex > maxIndex):
695 695 raise ValueError, "some value in (%d,%d) is not valid" % (minIndex, maxIndex)
696 696
697 697 if (maxIndex >= self.dataOut.nHeights):
698 698 maxIndex = self.dataOut.nHeights-1
699 699
700 700 # seleccion de indices para velocidades
701 701 indminvel = numpy.where(velrange >= minVel)
702 702 indmaxvel = numpy.where(velrange <= maxVel)
703 703 try:
704 704 minIndexVel = indminvel[0][0]
705 705 except:
706 706 minIndexVel = 0
707 707
708 708 try:
709 709 maxIndexVel = indmaxvel[0][-1]
710 710 except:
711 711 maxIndexVel = len(velrange)
712 712
713 713 #seleccion del espectro
714 714 data_spc = self.dataOut.data_spc[:,minIndexVel:maxIndexVel+1,minIndex:maxIndex+1]
715 715 #estimacion de ruido
716 716 noise = numpy.zeros(self.dataOut.nChannels)
717 717
718 718 for channel in range(self.dataOut.nChannels):
719 719 daux = data_spc[channel,:,:]
720 720 noise[channel] = hildebrand_sekhon(daux, self.dataOut.nIncohInt)
721 721
722 722 self.dataOut.noise_estimation = noise.copy()
723 723
724 724 return 1
725 725
726 726 class IncohInt(Operation):
727 727
728 728
729 729 __profIndex = 0
730 730 __withOverapping = False
731 731
732 732 __byTime = False
733 733 __initime = None
734 734 __lastdatatime = None
735 735 __integrationtime = None
736 736
737 737 __buffer_spc = None
738 738 __buffer_cspc = None
739 739 __buffer_dc = None
740 740
741 741 __dataReady = False
742 742
743 743 __timeInterval = None
744 744
745 745 n = None
746 746
747 747
748 748
749 749 def __init__(self):
750 750
751 751 Operation.__init__(self)
752 752 # self.isConfig = False
753 753
754 754 def setup(self, n=None, timeInterval=None, overlapping=False):
755 755 """
756 756 Set the parameters of the integration class.
757 757
758 758 Inputs:
759 759
760 760 n : Number of coherent integrations
761 761 timeInterval : Time of integration. If the parameter "n" is selected this one does not work
762 762 overlapping :
763 763
764 764 """
765 765
766 766 self.__initime = None
767 767 self.__lastdatatime = 0
768 768
769 769 self.__buffer_spc = 0
770 770 self.__buffer_cspc = 0
771 771 self.__buffer_dc = 0
772 772
773 773 self.__profIndex = 0
774 774 self.__dataReady = False
775 775 self.__byTime = False
776 776
777 777 if n is None and timeInterval is None:
778 778 raise ValueError, "n or timeInterval should be specified ..."
779 779
780 780 if n is not None:
781 781 self.n = int(n)
782 782 else:
783 783 self.__integrationtime = int(timeInterval) #if (type(timeInterval)!=integer) -> change this line
784 784 self.n = None
785 785 self.__byTime = True
786 786
787 787 def putData(self, data_spc, data_cspc, data_dc):
788 788
789 789 """
790 790 Add a profile to the __buffer_spc and increase in one the __profileIndex
791 791
792 792 """
793 793
794 794 self.__buffer_spc += data_spc
795 795
796 796 if data_cspc is None:
797 797 self.__buffer_cspc = None
798 798 else:
799 799 self.__buffer_cspc += data_cspc
800 800
801 801 if data_dc is None:
802 802 self.__buffer_dc = None
803 803 else:
804 804 self.__buffer_dc += data_dc
805 805
806 806 self.__profIndex += 1
807 807
808 808 return
809 809
810 810 def pushData(self):
811 811 """
812 812 Return the sum of the last profiles and the profiles used in the sum.
813 813
814 814 Affected:
815 815
816 816 self.__profileIndex
817 817
818 818 """
819 819
820 820 data_spc = self.__buffer_spc
821 821 data_cspc = self.__buffer_cspc
822 822 data_dc = self.__buffer_dc
823 823 n = self.__profIndex
824 824
825 825 self.__buffer_spc = 0
826 826 self.__buffer_cspc = 0
827 827 self.__buffer_dc = 0
828 828 self.__profIndex = 0
829 829
830 830 return data_spc, data_cspc, data_dc, n
831 831
832 832 def byProfiles(self, *args):
833 833
834 834 self.__dataReady = False
835 835 avgdata_spc = None
836 836 avgdata_cspc = None
837 837 avgdata_dc = None
838 838
839 839 self.putData(*args)
840 840
841 841 if self.__profIndex == self.n:
842 842
843 843 avgdata_spc, avgdata_cspc, avgdata_dc, n = self.pushData()
844 844 self.n = n
845 845 self.__dataReady = True
846 846
847 847 return avgdata_spc, avgdata_cspc, avgdata_dc
848 848
849 849 def byTime(self, datatime, *args):
850 850
851 851 self.__dataReady = False
852 852 avgdata_spc = None
853 853 avgdata_cspc = None
854 854 avgdata_dc = None
855 855
856 856 self.putData(*args)
857 857
858 858 if (datatime - self.__initime) >= self.__integrationtime:
859 859 avgdata_spc, avgdata_cspc, avgdata_dc, n = self.pushData()
860 860 self.n = n
861 861 self.__dataReady = True
862 862
863 863 return avgdata_spc, avgdata_cspc, avgdata_dc
864 864
865 865 def integrate(self, datatime, *args):
866 866
867 867 if self.__profIndex == 0:
868 868 self.__initime = datatime
869 869
870 870 if self.__byTime:
871 871 avgdata_spc, avgdata_cspc, avgdata_dc = self.byTime(datatime, *args)
872 872 else:
873 873 avgdata_spc, avgdata_cspc, avgdata_dc = self.byProfiles(*args)
874 874
875 875 if not self.__dataReady:
876 876 return None, None, None, None
877 877
878 878 return self.__initime, avgdata_spc, avgdata_cspc, avgdata_dc
879 879
880 880 def run(self, dataOut, n=None, timeInterval=None, overlapping=False):
881 881
882 882 if n==1:
883 883 return
884 884
885 885 dataOut.flagNoData = True
886 886
887 887 if not self.isConfig:
888 888 self.setup(n, timeInterval, overlapping)
889 889 self.isConfig = True
890 890
891 891 avgdatatime, avgdata_spc, avgdata_cspc, avgdata_dc = self.integrate(dataOut.utctime,
892 892 dataOut.data_spc,
893 893 dataOut.data_cspc,
894 894 dataOut.data_dc)
895 895
896 896 if self.__dataReady:
897 897
898 898 dataOut.data_spc = avgdata_spc
899 899 dataOut.data_cspc = avgdata_cspc
900 900 dataOut.data_dc = avgdata_dc
901 901
902 902 dataOut.nIncohInt *= self.n
903 903 dataOut.utctime = avgdatatime
904 904 dataOut.flagNoData = False
General Comments 0
You need to be logged in to leave comments. Login now