| This diff has been collapsed as it changes many lines, (3402 lines changed) Show them Hide them | |||
| @@ -111,39 +111,48 class ParametersProc(ProcessingUnit): | |||
|
|
111 | 111 | self.__updateObjFromInput() |
|
|
112 | 112 | self.dataOut.utctimeInit = self.dataIn.utctime |
|
|
113 | 113 | self.dataOut.paramInterval = self.dataIn.timeInterval |
|
|
114 | ||
|
|
115 | return | |
|
|
114 | 116 | |
|
|
115 | #------------------- Get Moments ---------------------------------- | |
|
|
116 | def GetMoments(self, channelList = None): | |
|
|
117 |
|
|
|
|
118 |
Function |
|
|
|
117 | class SpectralMoments(Operation): | |
|
|
118 | ||
|
|
119 | ''' | |
|
|
120 | Function SpectralMoments() | |
|
|
121 | ||
|
|
122 | Calculates moments (power, mean, standard deviation) and SNR of the signal | |
|
|
123 | ||
|
|
124 | Type of dataIn: Spectra | |
|
|
119 | 125 | |
|
|
120 | 126 | Input: |
|
|
121 | 127 | channelList : simple channel list to select e.g. [2,3,7] |
|
|
122 | self.dataOut.data_pre | |
|
|
123 | self.dataOut.abscissaList | |
|
|
124 | self.dataOut.noise | |
|
|
128 | self.dataOut.data_pre : Spectral data | |
|
|
129 | self.dataOut.abscissaList : List of frequencies | |
|
|
130 | self.dataOut.noise : Noise level per channel | |
|
|
125 | 131 | |
|
|
126 | 132 | Affected: |
|
|
127 | self.dataOut.data_param | |
|
|
128 | self.dataOut.data_SNR | |
|
|
133 | self.dataOut.data_param : Parameters per channel | |
|
|
134 | self.dataOut.data_SNR : SNR per channel | |
|
|
129 | 135 | |
|
|
130 |
|
|
|
|
131 | self.dataOut.data_pre = self.dataOut.data_pre[0] | |
|
|
132 | data = self.dataOut.data_pre | |
|
|
133 | absc = self.dataOut.abscissaList[:-1] | |
|
|
134 | noise = self.dataOut.noise | |
|
|
136 | ''' | |
|
|
137 | ||
|
|
138 | def run(self, dataOut, channelList=None): | |
|
|
139 | ||
|
|
140 | dataOut.data_pre = dataOut.data_pre[0] | |
|
|
141 | data = dataOut.data_pre | |
|
|
142 | absc = dataOut.abscissaList[:-1] | |
|
|
143 | noise = dataOut.noise | |
|
|
135 | 144 | |
|
|
136 | 145 | data_param = numpy.zeros((data.shape[0], 4, data.shape[2])) |
|
|
137 | 146 | |
|
|
138 | 147 | if channelList== None: |
|
|
139 |
channelList = |
|
|
|
140 |
|
|
|
|
148 | channelList = dataOut.channelList | |
|
|
149 | dataOut.channelList = channelList | |
|
|
141 | 150 | |
|
|
142 | 151 | for ind in channelList: |
|
|
143 | 152 | data_param[ind,:,:] = self.__calculateMoments(data[ind,:,:], absc, noise[ind]) |
|
|
144 | 153 | |
|
|
145 |
|
|
|
|
146 |
|
|
|
|
154 | dataOut.data_param = data_param[:,1:,:] | |
|
|
155 | dataOut.data_SNR = data_param[:,0] | |
|
|
147 | 156 | return |
|
|
148 | 157 | |
|
|
149 | 158 | def __calculateMoments(self, oldspec, oldfreq, n0, nicoh = None, graph = None, smooth = None, type1 = None, fwindow = None, snrth = None, dc = None, aliasing = None, oldfd = None, wwauto = None): |
| @@ -221,6 +230,7 class ParametersProc(ProcessingUnit): | |||
|
|
221 | 230 | #------------------ Get SA Parameters -------------------------- |
|
|
222 | 231 | |
|
|
223 | 232 | def GetSAParameters(self): |
|
|
233 | #SA en frecuencia | |
|
|
224 | 234 | pairslist = self.dataOut.groupList |
|
|
225 | 235 | num_pairs = len(pairslist) |
|
|
226 | 236 | |
| @@ -249,61 +259,64 class ParametersProc(ProcessingUnit): | |||
|
|
249 | 259 | cspc_par[i,:,:] = __calculateMoments(norm_cspectra) |
|
|
250 | 260 | #------------------- Get Lags ---------------------------------- |
|
|
251 | 261 | |
|
|
252 | def GetLags(self): | |
|
|
253 |
|
|
|
|
254 |
|
|
|
|
262 | class SALags(Operation): | |
|
|
263 | ''' | |
|
|
264 | Function GetMoments() | |
|
|
255 | 265 | |
|
|
256 |
|
|
|
|
257 |
|
|
|
|
258 |
|
|
|
|
259 |
|
|
|
|
260 |
|
|
|
|
261 |
|
|
|
|
262 |
|
|
|
|
263 |
|
|
|
|
264 |
|
|
|
|
265 |
|
|
|
|
266 |
|
|
|
|
267 |
|
|
|
|
268 |
|
|
|
|
269 | ||
|
|
270 |
data = |
|
|
|
271 | normFactor = self.dataOut.normFactor | |
|
|
272 | nHeights = self.dataOut.nHeights | |
|
|
273 | absc = self.dataOut.abscissaList[:-1] | |
|
|
274 |
no |
|
|
|
275 | SNR = self.dataOut.data_SNR | |
|
|
276 |
|
|
|
|
277 |
n |
|
|
|
278 | pairsAutoCorr, pairsCrossCorr = self.__getPairsAutoCorr(pairsList, nChannels) | |
|
|
279 | self.dataOut.data_param = numpy.zeros((len(pairsCrossCorr)*2 + 1, nHeights)) | |
|
|
266 | Input: | |
|
|
267 | self.dataOut.data_pre | |
|
|
268 | self.dataOut.abscissaList | |
|
|
269 | self.dataOut.noise | |
|
|
270 | self.dataOut.normFactor | |
|
|
271 | self.dataOut.data_SNR | |
|
|
272 | self.dataOut.groupList | |
|
|
273 | self.dataOut.nChannels | |
|
|
274 | ||
|
|
275 | Affected: | |
|
|
276 | self.dataOut.data_param | |
|
|
277 | ||
|
|
278 | ''' | |
|
|
279 | def run(self, dataOut): | |
|
|
280 | data = dataOut.data_pre | |
|
|
281 | data_acf = dataOut.data_pre[0] | |
|
|
282 | data_ccf = dataOut.data_pre[1] | |
|
|
283 | ||
|
|
284 | normFactor = dataOut.normFactor | |
|
|
285 | nHeights = dataOut.nHeights | |
|
|
286 | absc = dataOut.abscissaList[:-1] | |
|
|
287 | noise = dataOut.noise | |
|
|
288 | SNR = dataOut.data_SNR | |
|
|
289 | # pairsList = dataOut.groupList | |
|
|
290 | nChannels = dataOut.nChannels | |
|
|
291 | # pairsAutoCorr, pairsCrossCorr = self.__getPairsAutoCorr(pairsList, nChannels) | |
|
|
292 | dataOut.data_param = numpy.zeros((len(pairsCrossCorr)*2 + 1, nHeights)) | |
|
|
280 | 293 | |
|
|
281 | 294 | dataNorm = numpy.abs(data) |
|
|
282 | 295 | for l in range(len(pairsList)): |
|
|
283 | 296 | dataNorm[l,:,:] = dataNorm[l,:,:]/normFactor[l,:] |
|
|
284 | 297 | |
|
|
285 | 298 | self.dataOut.data_param[:-1,:] = self.__calculateTaus(dataNorm, pairsCrossCorr, pairsAutoCorr, absc) |
|
|
286 |
self.dataOut.data_param[-1,:] = self.__calculateLag1Phase(data |
|
|
|
299 | self.dataOut.data_param[-1,:] = self.__calculateLag1Phase(data_acf, absc) | |
|
|
287 | 300 | return |
|
|
288 | 301 | |
|
|
289 | def __getPairsAutoCorr(self, pairsList, nChannels): | |
|
|
290 | ||
|
|
291 | pairsAutoCorr = numpy.zeros(nChannels, dtype = 'int')*numpy.nan | |
|
|
292 | ||
|
|
293 | for l in range(len(pairsList)): | |
|
|
294 | firstChannel = pairsList[l][0] | |
|
|
295 | secondChannel = pairsList[l][1] | |
|
|
296 | ||
|
|
297 | #Obteniendo pares de Autocorrelacion | |
|
|
298 | if firstChannel == secondChannel: | |
|
|
299 | pairsAutoCorr[firstChannel] = int(l) | |
|
|
300 | ||
|
|
301 | pairsAutoCorr = pairsAutoCorr.astype(int) | |
|
|
302 | ||
|
|
303 | pairsCrossCorr = range(len(pairsList)) | |
|
|
304 | pairsCrossCorr = numpy.delete(pairsCrossCorr,pairsAutoCorr) | |
|
|
305 | ||
|
|
306 | return pairsAutoCorr, pairsCrossCorr | |
|
|
302 | # def __getPairsAutoCorr(self, pairsList, nChannels): | |
|
|
303 | # | |
|
|
304 | # pairsAutoCorr = numpy.zeros(nChannels, dtype = 'int')*numpy.nan | |
|
|
305 | # | |
|
|
306 | # for l in range(len(pairsList)): | |
|
|
307 | # firstChannel = pairsList[l][0] | |
|
|
308 | # secondChannel = pairsList[l][1] | |
|
|
309 | # | |
|
|
310 | # #Obteniendo pares de Autocorrelacion | |
|
|
311 | # if firstChannel == secondChannel: | |
|
|
312 | # pairsAutoCorr[firstChannel] = int(l) | |
|
|
313 | # | |
|
|
314 | # pairsAutoCorr = pairsAutoCorr.astype(int) | |
|
|
315 | # | |
|
|
316 | # pairsCrossCorr = range(len(pairsList)) | |
|
|
317 | # pairsCrossCorr = numpy.delete(pairsCrossCorr,pairsAutoCorr) | |
|
|
318 | # | |
|
|
319 | # return pairsAutoCorr, pairsCrossCorr | |
|
|
307 | 320 | |
|
|
308 | 321 | def __calculateTaus(self, data, pairsCrossCorr, pairsAutoCorr, lagTRange): |
|
|
309 | 322 | |
| @@ -331,1016 +344,913 class ParametersProc(ProcessingUnit): | |||
|
|
331 | 344 | |
|
|
332 | 345 | return tau |
|
|
333 | 346 | |
|
|
334 |
def __calculateLag1Phase(self, data, |
|
|
|
335 |
data1 = stats.nanmean(data |
|
|
|
347 | def __calculateLag1Phase(self, data, lagTRange): | |
|
|
348 | data1 = stats.nanmean(data, axis = 0) | |
|
|
336 | 349 | lag1 = numpy.where(lagTRange == 0)[0][0] + 1 |
|
|
337 | 350 | |
|
|
338 | 351 | phase = numpy.angle(data1[lag1,:]) |
|
|
339 | 352 | |
|
|
340 | 353 | return phase |
|
|
341 | #------------------- Detect Meteors ------------------------------ | |
|
|
342 | 354 | |
|
|
343 | def MeteorDetection(self, hei_ref = None, tauindex = 0, | |
|
|
344 | phaseOffsets = None, | |
|
|
345 | cohDetection = False, cohDet_timeStep = 1, cohDet_thresh = 25, | |
|
|
346 | noise_timeStep = 4, noise_multiple = 4, | |
|
|
347 | multDet_timeLimit = 1, multDet_rangeLimit = 3, | |
|
|
348 | phaseThresh = 20, SNRThresh = 5, | |
|
|
349 | hmin = 50, hmax=150, azimuth = 0, | |
|
|
350 | # channel25X = 0, channel20X = 4, channelCentX = 2, | |
|
|
351 | # channel25Y = 1, channel20Y = 3, channelCentY = 2, | |
|
|
352 | channelPositions = None) : | |
|
|
355 | class SpectralFitting(Operation): | |
|
|
356 | ''' | |
|
|
357 | Function GetMoments() | |
|
|
353 | 358 |
|
|
|
354 | ''' | |
|
|
355 | Function DetectMeteors() | |
|
|
356 | Project developed with paper: | |
|
|
357 | HOLDSWORTH ET AL. 2004 | |
|
|
358 | ||
|
|
359 | 359 | Input: |
|
|
360 | self.dataOut.data_pre | |
|
|
361 | ||
|
|
362 | centerReceiverIndex: From the channels, which is the center receiver | |
|
|
363 | ||
|
|
364 | hei_ref: Height reference for the Beacon signal extraction | |
|
|
365 | tauindex: | |
|
|
366 | predefinedPhaseShifts: Predefined phase offset for the voltge signals | |
|
|
367 | ||
|
|
368 | cohDetection: Whether to user Coherent detection or not | |
|
|
369 | cohDet_timeStep: Coherent Detection calculation time step | |
|
|
370 | cohDet_thresh: Coherent Detection phase threshold to correct phases | |
|
|
371 | ||
|
|
372 | noise_timeStep: Noise calculation time step | |
|
|
373 | noise_multiple: Noise multiple to define signal threshold | |
|
|
374 | ||
|
|
375 | multDet_timeLimit: Multiple Detection Removal time limit in seconds | |
|
|
376 | multDet_rangeLimit: Multiple Detection Removal range limit in km | |
|
|
377 | ||
|
|
378 | phaseThresh: Maximum phase difference between receiver to be consider a meteor | |
|
|
379 | SNRThresh: Minimum SNR threshold of the meteor signal to be consider a meteor | |
|
|
380 | ||
|
|
381 | hmin: Minimum Height of the meteor to use it in the further wind estimations | |
|
|
382 | hmax: Maximum Height of the meteor to use it in the further wind estimations | |
|
|
383 | azimuth: Azimuth angle correction | |
|
|
384 | ||
|
|
385 | Affected: | |
|
|
386 | self.dataOut.data_param | |
|
|
387 | ||
|
|
388 | Rejection Criteria (Errors): | |
|
|
389 | 0: No error; analysis OK | |
|
|
390 | 1: SNR < SNR threshold | |
|
|
391 | 2: angle of arrival (AOA) ambiguously determined | |
|
|
392 | 3: AOA estimate not feasible | |
|
|
393 | 4: Large difference in AOAs obtained from different antenna baselines | |
|
|
394 | 5: echo at start or end of time series | |
|
|
395 | 6: echo less than 5 examples long; too short for analysis | |
|
|
396 | 7: echo rise exceeds 0.3s | |
|
|
397 | 8: echo decay time less than twice rise time | |
|
|
398 | 9: large power level before echo | |
|
|
399 | 10: large power level after echo | |
|
|
400 | 11: poor fit to amplitude for estimation of decay time | |
|
|
401 | 12: poor fit to CCF phase variation for estimation of radial drift velocity | |
|
|
402 | 13: height unresolvable echo: not valid height within 70 to 110 km | |
|
|
403 | 14: height ambiguous echo: more then one possible height within 70 to 110 km | |
|
|
404 | 15: radial drift velocity or projected horizontal velocity exceeds 200 m/s | |
|
|
405 | 16: oscilatory echo, indicating event most likely not an underdense echo | |
|
|
406 | ||
|
|
407 | 17: phase difference in meteor Reestimation | |
|
|
408 | ||
|
|
409 | Data Storage: | |
|
|
410 | Meteors for Wind Estimation (8): | |
|
|
411 | Utc Time | Range Height | |
|
|
412 | Azimuth Zenith errorCosDir | |
|
|
413 | VelRad errorVelRad | |
|
|
414 | Phase0 Phase1 Phase2 Phase3 | |
|
|
415 | TypeError | |
|
|
416 | ||
|
|
417 | ''' | |
|
|
418 | #Getting Pairslist | |
|
|
419 | if channelPositions == None: | |
|
|
420 | # channelPositions = [(2.5,0), (0,2.5), (0,0), (0,4.5), (-2,0)] #T | |
|
|
421 | channelPositions = [(4.5,2), (2,4.5), (2,2), (2,0), (0,2)] #Estrella | |
|
|
422 | meteorOps = MeteorOperations() | |
|
|
423 | pairslist0, distances = meteorOps.getPhasePairs(channelPositions) | |
|
|
360 | Output: | |
|
|
361 | Variables modified: | |
|
|
362 | ''' | |
|
|
363 | ||
|
|
364 | def run(self, dataOut, getSNR = True, path=None, file=None, groupList=None): | |
|
|
424 | 365 | |
|
|
425 | #Get Beacon signal | |
|
|
426 | newheis = numpy.where(self.dataOut.heightList>self.dataOut.radarControllerHeaderObj.Taus[tauindex]) | |
|
|
427 | 366 | |
|
|
428 |
if h |
|
|
|
429 | newheis = numpy.where(self.dataOut.heightList>hei_ref) | |
|
|
367 | if path != None: | |
|
|
368 | sys.path.append(path) | |
|
|
369 | self.dataOut.library = importlib.import_module(file) | |
|
|
430 | 370 | |
|
|
431 | heiRang = self.dataOut.getHeiRange() | |
|
|
432 | ||
|
|
433 | # nChannel = self.dataOut.nChannels | |
|
|
434 | # for i in range(nChannel): | |
|
|
435 | # if i != centerReceiverIndex: | |
|
|
436 | # pairslist.append((centerReceiverIndex,i)) | |
|
|
437 | ||
|
|
438 | #****************REMOVING HARDWARE PHASE DIFFERENCES*************** | |
|
|
439 | # see if the user put in pre defined phase shifts | |
|
|
440 | voltsPShift = self.dataOut.data_pre.copy() | |
|
|
371 | #To be inserted as a parameter | |
|
|
372 | groupArray = numpy.array(groupList) | |
|
|
373 | # groupArray = numpy.array([[0,1],[2,3]]) | |
|
|
374 | self.dataOut.groupList = groupArray | |
|
|
441 | 375 | |
|
|
442 | # if predefinedPhaseShifts != None: | |
|
|
443 | # hardwarePhaseShifts = numpy.array(predefinedPhaseShifts)*numpy.pi/180 | |
|
|
444 | # | |
|
|
445 | # # elif beaconPhaseShifts: | |
|
|
446 | # # #get hardware phase shifts using beacon signal | |
|
|
447 | # # hardwarePhaseShifts = self.__getHardwarePhaseDiff(self.dataOut.data_pre, pairslist, newheis, 10) | |
|
|
448 | # # hardwarePhaseShifts = numpy.insert(hardwarePhaseShifts,centerReceiverIndex,0) | |
|
|
449 | # | |
|
|
450 | # else: | |
|
|
451 | # hardwarePhaseShifts = numpy.zeros(5) | |
|
|
452 | # | |
|
|
453 | # voltsPShift = numpy.zeros((self.dataOut.data_pre.shape[0],self.dataOut.data_pre.shape[1],self.dataOut.data_pre.shape[2]), dtype = 'complex') | |
|
|
454 | # for i in range(self.dataOut.data_pre.shape[0]): | |
|
|
455 | # voltsPShift[i,:,:] = self.__shiftPhase(self.dataOut.data_pre[i,:,:], hardwarePhaseShifts[i]) | |
|
|
456 | ||
|
|
457 | #******************END OF REMOVING HARDWARE PHASE DIFFERENCES********* | |
|
|
458 | ||
|
|
459 | #Remove DC | |
|
|
460 | voltsDC = numpy.mean(voltsPShift,1) | |
|
|
461 | voltsDC = numpy.mean(voltsDC,1) | |
|
|
462 | for i in range(voltsDC.shape[0]): | |
|
|
463 | voltsPShift[i] = voltsPShift[i] - voltsDC[i] | |
|
|
464 | ||
|
|
465 | #Don't considerate last heights, theyre used to calculate Hardware Phase Shift | |
|
|
466 | voltsPShift = voltsPShift[:,:,:newheis[0][0]] | |
|
|
376 | nGroups = groupArray.shape[0] | |
|
|
377 | nChannels = self.dataIn.nChannels | |
|
|
378 | nHeights=self.dataIn.heightList.size | |
|
|
467 | 379 | |
|
|
468 | #************ FIND POWER OF DATA W/COH OR NON COH DETECTION (3.4) ********** | |
|
|
469 | #Coherent Detection | |
|
|
470 | if cohDetection: | |
|
|
471 | #use coherent detection to get the net power | |
|
|
472 | cohDet_thresh = cohDet_thresh*numpy.pi/180 | |
|
|
473 | voltsPShift = self.__coherentDetection(voltsPShift, cohDet_timeStep, self.dataOut.timeInterval, pairslist0, cohDet_thresh) | |
|
|
380 | #Parameters Array | |
|
|
381 | self.dataOut.data_param = None | |
|
|
474 | 382 | |
|
|
475 | #Non-coherent detection! | |
|
|
476 | powerNet = numpy.nansum(numpy.abs(voltsPShift[:,:,:])**2,0) | |
|
|
477 | #********** END OF COH/NON-COH POWER CALCULATION********************** | |
|
|
478 | ||
|
|
479 | #********** FIND THE NOISE LEVEL AND POSSIBLE METEORS **************** | |
|
|
480 | #Get noise | |
|
|
481 | noise, noise1 = self.__getNoise(powerNet, noise_timeStep, self.dataOut.timeInterval) | |
|
|
482 | # noise = self.getNoise1(powerNet, noise_timeStep, self.dataOut.timeInterval) | |
|
|
483 | #Get signal threshold | |
|
|
484 | signalThresh = noise_multiple*noise | |
|
|
485 | #Meteor echoes detection | |
|
|
486 | listMeteors = self.__findMeteors(powerNet, signalThresh) | |
|
|
487 | #******* END OF NOISE LEVEL AND POSSIBLE METEORS CACULATION ********** | |
|
|
383 | #Set constants | |
|
|
384 | constants = self.dataOut.library.setConstants(self.dataIn) | |
|
|
385 | self.dataOut.constants = constants | |
|
|
386 | M = self.dataIn.normFactor | |
|
|
387 | N = self.dataIn.nFFTPoints | |
|
|
388 | ippSeconds = self.dataIn.ippSeconds | |
|
|
389 | K = self.dataIn.nIncohInt | |
|
|
390 | pairsArray = numpy.array(self.dataIn.pairsList) | |
|
|
488 | 391 | |
|
|
489 | #************** REMOVE MULTIPLE DETECTIONS (3.5) *************************** | |
|
|
490 | #Parameters | |
|
|
491 | heiRange = self.dataOut.getHeiRange() | |
|
|
492 | rangeInterval = heiRange[1] - heiRange[0] | |
|
|
493 | rangeLimit = multDet_rangeLimit/rangeInterval | |
|
|
494 | timeLimit = multDet_timeLimit/self.dataOut.timeInterval | |
|
|
495 | #Multiple detection removals | |
|
|
496 | listMeteors1 = self.__removeMultipleDetections(listMeteors, rangeLimit, timeLimit) | |
|
|
497 | #************ END OF REMOVE MULTIPLE DETECTIONS ********************** | |
|
|
392 | #List of possible combinations | |
|
|
393 | listComb = itertools.combinations(numpy.arange(groupArray.shape[1]),2) | |
|
|
394 | indCross = numpy.zeros(len(list(listComb)), dtype = 'int') | |
|
|
498 | 395 | |
|
|
499 | #********************* METEOR REESTIMATION (3.7, 3.8, 3.9, 3.10) ******************** | |
|
|
500 | #Parameters | |
|
|
501 | phaseThresh = phaseThresh*numpy.pi/180 | |
|
|
502 | thresh = [phaseThresh, noise_multiple, SNRThresh] | |
|
|
503 | #Meteor reestimation (Errors N 1, 6, 12, 17) | |
|
|
504 | listMeteors2, listMeteorsPower, listMeteorsVolts = self.__meteorReestimation(listMeteors1, voltsPShift, pairslist0, thresh, noise, self.dataOut.timeInterval, self.dataOut.frequency) | |
|
|
505 | # listMeteors2, listMeteorsPower, listMeteorsVolts = self.meteorReestimation3(listMeteors2, listMeteorsPower, listMeteorsVolts, voltsPShift, pairslist, thresh, noise) | |
|
|
506 | #Estimation of decay times (Errors N 7, 8, 11) | |
|
|
507 | listMeteors3 = self.__estimateDecayTime(listMeteors2, listMeteorsPower, self.dataOut.timeInterval, self.dataOut.frequency) | |
|
|
508 | #******************* END OF METEOR REESTIMATION ******************* | |
|
|
396 | if getSNR: | |
|
|
397 | listChannels = groupArray.reshape((groupArray.size)) | |
|
|
398 | listChannels.sort() | |
|
|
399 | noise = self.dataIn.getNoise() | |
|
|
400 | self.dataOut.data_SNR = self.__getSNR(self.dataIn.data_spc[listChannels,:,:], noise[listChannels]) | |
|
|
509 | 401 | |
|
|
510 | #********************* METEOR PARAMETERS CALCULATION (3.11, 3.12, 3.13) ************************** | |
|
|
511 | #Calculating Radial Velocity (Error N 15) | |
|
|
512 | radialStdThresh = 10 | |
|
|
513 | listMeteors4 = self.__getRadialVelocity(listMeteors3, listMeteorsVolts, radialStdThresh, pairslist0, self.dataOut.timeInterval) | |
|
|
514 | ||
|
|
515 | if len(listMeteors4) > 0: | |
|
|
516 | #Setting New Array | |
|
|
517 | date = self.dataOut.utctime | |
|
|
518 | arrayParameters = self.__setNewArrays(listMeteors4, date, heiRang) | |
|
|
519 | ||
|
|
520 | #Correcting phase offset | |
|
|
521 | if phaseOffsets != None: | |
|
|
522 | phaseOffsets = numpy.array(phaseOffsets)*numpy.pi/180 | |
|
|
523 | arrayParameters[:,8:12] = numpy.unwrap(arrayParameters[:,8:12] + phaseOffsets) | |
|
|
402 | for i in range(nGroups): | |
|
|
403 | coord = groupArray[i,:] | |
|
|
524 | 404 | |
|
|
525 |
# |
|
|
|
526 | pairsList = [] | |
|
|
527 | pairx = (0,1) | |
|
|
528 | pairy = (2,3) | |
|
|
529 | pairsList.append(pairx) | |
|
|
530 | pairsList.append(pairy) | |
|
|
405 | #Input data array | |
|
|
406 | data = self.dataIn.data_spc[coord,:,:]/(M*N) | |
|
|
407 | data = data.reshape((data.shape[0]*data.shape[1],data.shape[2])) | |
|
|
531 | 408 | |
|
|
532 | jph = numpy.array([0,0,0,0]) | |
|
|
533 |
|
|
|
|
534 | arrayParameters = meteorOps.getMeteorParams(arrayParameters, azimuth, h, pairsList, distances, jph) | |
|
|
409 | #Cross Spectra data array for Covariance Matrixes | |
|
|
410 | ind = 0 | |
|
|
411 | for pairs in listComb: | |
|
|
412 | pairsSel = numpy.array([coord[x],coord[y]]) | |
|
|
413 | indCross[ind] = int(numpy.where(numpy.all(pairsArray == pairsSel, axis = 1))[0][0]) | |
|
|
414 | ind += 1 | |
|
|
415 | dataCross = self.dataIn.data_cspc[indCross,:,:]/(M*N) | |
|
|
416 | dataCross = dataCross**2/K | |
|
|
535 | 417 | |
|
|
536 | # #Calculate AOA (Error N 3, 4) | |
|
|
537 | # #JONES ET AL. 1998 | |
|
|
538 | # error = arrayParameters[:,-1] | |
|
|
539 | # AOAthresh = numpy.pi/8 | |
|
|
540 | # phases = -arrayParameters[:,9:13] | |
|
|
541 | # arrayParameters[:,4:7], arrayParameters[:,-1] = meteorOps.getAOA(phases, pairsList, error, AOAthresh, azimuth) | |
|
|
542 | # | |
|
|
543 | # #Calculate Heights (Error N 13 and 14) | |
|
|
544 | # error = arrayParameters[:,-1] | |
|
|
545 | # Ranges = arrayParameters[:,2] | |
|
|
546 | # zenith = arrayParameters[:,5] | |
|
|
547 | # arrayParameters[:,3], arrayParameters[:,-1] = meteorOps.getHeights(Ranges, zenith, error, hmin, hmax) | |
|
|
548 | # error = arrayParameters[:,-1] | |
|
|
549 | #********************* END OF PARAMETERS CALCULATION ************************** | |
|
|
418 | for h in range(nHeights): | |
|
|
419 | # print self.dataOut.heightList[h] | |
|
|
550 | 420 | |
|
|
551 | #***************************+ PASS DATA TO NEXT STEP ********************** | |
|
|
552 | # arrayFinal = arrayParameters.reshape((1,arrayParameters.shape[0],arrayParameters.shape[1])) | |
|
|
553 | self.dataOut.data_param = arrayParameters | |
|
|
554 | ||
|
|
555 | if arrayParameters == None: | |
|
|
556 | self.dataOut.flagNoData = True | |
|
|
557 | else: | |
|
|
558 | self.dataOut.flagNoData = True | |
|
|
559 | ||
|
|
421 | #Input | |
|
|
422 | d = data[:,h] | |
|
|
423 | ||
|
|
424 | #Covariance Matrix | |
|
|
425 | D = numpy.diag(d**2/K) | |
|
|
426 | ind = 0 | |
|
|
427 | for pairs in listComb: | |
|
|
428 | #Coordinates in Covariance Matrix | |
|
|
429 | x = pairs[0] | |
|
|
430 | y = pairs[1] | |
|
|
431 | #Channel Index | |
|
|
432 | S12 = dataCross[ind,:,h] | |
|
|
433 | D12 = numpy.diag(S12) | |
|
|
434 | #Completing Covariance Matrix with Cross Spectras | |
|
|
435 | D[x*N:(x+1)*N,y*N:(y+1)*N] = D12 | |
|
|
436 | D[y*N:(y+1)*N,x*N:(x+1)*N] = D12 | |
|
|
437 | ind += 1 | |
|
|
438 | Dinv=numpy.linalg.inv(D) | |
|
|
439 | L=numpy.linalg.cholesky(Dinv) | |
|
|
440 | LT=L.T | |
|
|
441 | ||
|
|
442 | dp = numpy.dot(LT,d) | |
|
|
443 | ||
|
|
444 | #Initial values | |
|
|
445 | data_spc = self.dataIn.data_spc[coord,:,h] | |
|
|
446 | ||
|
|
447 | if (h>0)and(error1[3]<5): | |
|
|
448 | p0 = self.dataOut.data_param[i,:,h-1] | |
|
|
449 | else: | |
|
|
450 | p0 = numpy.array(self.dataOut.library.initialValuesFunction(data_spc, constants, i)) | |
|
|
451 | ||
|
|
452 | try: | |
|
|
453 | #Least Squares | |
|
|
454 | minp,covp,infodict,mesg,ier = optimize.leastsq(self.__residFunction,p0,args=(dp,LT,constants),full_output=True) | |
|
|
455 | # minp,covp = optimize.leastsq(self.__residFunction,p0,args=(dp,LT,constants)) | |
|
|
456 | #Chi square error | |
|
|
457 | error0 = numpy.sum(infodict['fvec']**2)/(2*N) | |
|
|
458 | #Error with Jacobian | |
|
|
459 | error1 = self.dataOut.library.errorFunction(minp,constants,LT) | |
|
|
460 | except: | |
|
|
461 | minp = p0*numpy.nan | |
|
|
462 | error0 = numpy.nan | |
|
|
463 | error1 = p0*numpy.nan | |
|
|
464 | ||
|
|
465 | #Save | |
|
|
466 | if self.dataOut.data_param == None: | |
|
|
467 | self.dataOut.data_param = numpy.zeros((nGroups, p0.size, nHeights))*numpy.nan | |
|
|
468 | self.dataOut.data_error = numpy.zeros((nGroups, p0.size + 1, nHeights))*numpy.nan | |
|
|
469 | ||
|
|
470 | self.dataOut.data_error[i,:,h] = numpy.hstack((error0,error1)) | |
|
|
471 | self.dataOut.data_param[i,:,h] = minp | |
|
|
560 | 472 | return |
|
|
561 | 473 | |
|
|
562 | def CorrectMeteorPhases(self, phaseOffsets, hmin = 50, hmax = 150, azimuth = 45, channelPositions = None): | |
|
|
563 | ||
|
|
564 | arrayParameters = self.dataOut.data_param | |
|
|
565 | pairsList = [] | |
|
|
566 | pairx = (0,1) | |
|
|
567 | pairy = (2,3) | |
|
|
568 | pairsList.append(pairx) | |
|
|
569 | pairsList.append(pairy) | |
|
|
570 | jph = numpy.zeros(4) | |
|
|
571 | ||
|
|
572 | phaseOffsets = numpy.array(phaseOffsets)*numpy.pi/180 | |
|
|
573 | # arrayParameters[:,8:12] = numpy.unwrap(arrayParameters[:,8:12] + phaseOffsets) | |
|
|
574 | arrayParameters[:,8:12] = numpy.angle(numpy.exp(1j*(arrayParameters[:,8:12] + phaseOffsets))) | |
|
|
575 | ||
|
|
576 | meteorOps = MeteorOperations() | |
|
|
577 | if channelPositions == None: | |
|
|
578 | # channelPositions = [(2.5,0), (0,2.5), (0,0), (0,4.5), (-2,0)] #T | |
|
|
579 | channelPositions = [(4.5,2), (2,4.5), (2,2), (2,0), (0,2)] #Estrella | |
|
|
580 | ||
|
|
581 | pairslist0, distances = meteorOps.getPhasePairs(channelPositions) | |
|
|
582 | h = (hmin,hmax) | |
|
|
474 | def __residFunction(self, p, dp, LT, constants): | |
|
|
583 | 475 | |
|
|
584 | arrayParameters = meteorOps.getMeteorParams(arrayParameters, azimuth, h, pairsList, distances, jph) | |
|
|
476 | fm = self.dataOut.library.modelFunction(p, constants) | |
|
|
477 | fmp=numpy.dot(LT,fm) | |
|
|
478 | ||
|
|
479 | return dp-fmp | |
|
|
480 | ||
|
|
481 | def __getSNR(self, z, noise): | |
|
|
585 | 482 | |
|
|
586 | self.dataOut.data_param = arrayParameters | |
|
|
587 | return | |
|
|
483 | avg = numpy.average(z, axis=1) | |
|
|
484 | SNR = (avg.T-noise)/noise | |
|
|
485 | SNR = SNR.T | |
|
|
486 | return SNR | |
|
|
588 | 487 | |
|
|
589 | def __getHardwarePhaseDiff(self, voltage0, pairslist, newheis, n): | |
|
|
590 | ||
|
|
591 | minIndex = min(newheis[0]) | |
|
|
592 | maxIndex = max(newheis[0]) | |
|
|
593 | ||
|
|
594 | voltage = voltage0[:,:,minIndex:maxIndex+1] | |
|
|
595 | nLength = voltage.shape[1]/n | |
|
|
596 | nMin = 0 | |
|
|
597 | nMax = 0 | |
|
|
598 | phaseOffset = numpy.zeros((len(pairslist),n)) | |
|
|
599 | ||
|
|
600 | for i in range(n): | |
|
|
601 | nMax += nLength | |
|
|
602 | phaseCCF = -numpy.angle(self.__calculateCCF(voltage[:,nMin:nMax,:], pairslist, [0])) | |
|
|
603 | phaseCCF = numpy.mean(phaseCCF, axis = 2) | |
|
|
604 | phaseOffset[:,i] = phaseCCF.transpose() | |
|
|
605 | nMin = nMax | |
|
|
606 | # phaseDiff, phaseArrival = self.estimatePhaseDifference(voltage, pairslist) | |
|
|
607 | ||
|
|
608 | #Remove Outliers | |
|
|
609 | factor = 2 | |
|
|
610 | wt = phaseOffset - signal.medfilt(phaseOffset,(1,5)) | |
|
|
611 | dw = numpy.std(wt,axis = 1) | |
|
|
612 | dw = dw.reshape((dw.size,1)) | |
|
|
613 | ind = numpy.where(numpy.logical_or(wt>dw*factor,wt<-dw*factor)) | |
|
|
614 | phaseOffset[ind] = numpy.nan | |
|
|
615 | phaseOffset = stats.nanmean(phaseOffset, axis=1) | |
|
|
488 | def __chisq(p,chindex,hindex): | |
|
|
489 | #similar to Resid but calculates CHI**2 | |
|
|
490 | [LT,d,fm]=setupLTdfm(p,chindex,hindex) | |
|
|
491 | dp=numpy.dot(LT,d) | |
|
|
492 | fmp=numpy.dot(LT,fm) | |
|
|
493 | chisq=numpy.dot((dp-fmp).T,(dp-fmp)) | |
|
|
494 | return chisq | |
|
|
495 | ||
|
|
496 | class WindProfiler(Operation): | |
|
|
497 | ||
|
|
498 | __isConfig = False | |
|
|
616 | 499 | |
|
|
617 | return phaseOffset | |
|
|
500 | __initime = None | |
|
|
501 | __lastdatatime = None | |
|
|
502 | __integrationtime = None | |
|
|
618 | 503 | |
|
|
619 | def __shiftPhase(self, data, phaseShift): | |
|
|
620 | #this will shift the phase of a complex number | |
|
|
621 | dataShifted = numpy.abs(data) * numpy.exp((numpy.angle(data)+phaseShift)*1j) | |
|
|
622 | return dataShifted | |
|
|
504 | __buffer = None | |
|
|
623 | 505 | |
|
|
624 | def __estimatePhaseDifference(self, array, pairslist): | |
|
|
625 | nChannel = array.shape[0] | |
|
|
626 | nHeights = array.shape[2] | |
|
|
627 | numPairs = len(pairslist) | |
|
|
628 | # phaseCCF = numpy.zeros((nChannel, 5, nHeights)) | |
|
|
629 | phaseCCF = numpy.angle(self.__calculateCCF(array, pairslist, [-2,-1,0,1,2])) | |
|
|
506 | __dataReady = False | |
|
|
507 | ||
|
|
508 | __firstdata = None | |
|
|
509 | ||
|
|
510 | n = None | |
|
|
511 | ||
|
|
512 | def __init__(self): | |
|
|
513 | Operation.__init__(self) | |
|
|
514 | ||
|
|
515 | def __calculateCosDir(self, elev, azim): | |
|
|
516 | zen = (90 - elev)*numpy.pi/180 | |
|
|
517 | azim = azim*numpy.pi/180 | |
|
|
518 | cosDirX = numpy.sqrt((1-numpy.cos(zen)**2)/((1+numpy.tan(azim)**2))) | |
|
|
519 | cosDirY = numpy.sqrt(1-numpy.cos(zen)**2-cosDirX**2) | |
|
|
630 | 520 | |
|
|
631 | #Correct phases | |
|
|
632 | derPhaseCCF = phaseCCF[:,1:,:] - phaseCCF[:,0:-1,:] | |
|
|
633 | indDer = numpy.where(numpy.abs(derPhaseCCF) > numpy.pi) | |
|
|
521 | signX = numpy.sign(numpy.cos(azim)) | |
|
|
522 | signY = numpy.sign(numpy.sin(azim)) | |
|
|
634 | 523 | |
|
|
635 | if indDer[0].shape[0] > 0: | |
|
|
636 | for i in range(indDer[0].shape[0]): | |
|
|
637 | signo = -numpy.sign(derPhaseCCF[indDer[0][i],indDer[1][i],indDer[2][i]]) | |
|
|
638 | phaseCCF[indDer[0][i],indDer[1][i]+1:,:] += signo*2*numpy.pi | |
|
|
524 | cosDirX = numpy.copysign(cosDirX, signX) | |
|
|
525 | cosDirY = numpy.copysign(cosDirY, signY) | |
|
|
526 | return cosDirX, cosDirY | |
|
|
639 | 527 | |
|
|
640 | # for j in range(numSides): | |
|
|
641 | # phaseCCFAux = self.calculateCCF(arrayCenter, arraySides[j,:,:], [-2,1,0,1,2]) | |
|
|
642 | # phaseCCF[j,:,:] = numpy.angle(phaseCCFAux) | |
|
|
643 | # | |
|
|
644 | #Linear | |
|
|
645 | phaseInt = numpy.zeros((numPairs,1)) | |
|
|
646 | angAllCCF = phaseCCF[:,[0,1,3,4],0] | |
|
|
647 | for j in range(numPairs): | |
|
|
648 | fit = stats.linregress([-2,-1,1,2],angAllCCF[j,:]) | |
|
|
649 | phaseInt[j] = fit[1] | |
|
|
650 | #Phase Differences | |
|
|
651 | phaseDiff = phaseInt - phaseCCF[:,2,:] | |
|
|
652 | phaseArrival = phaseInt.reshape(phaseInt.size) | |
|
|
528 | def __calculateAngles(self, theta_x, theta_y, azimuth): | |
|
|
529 | ||
|
|
530 | dir_cosw = numpy.sqrt(1-theta_x**2-theta_y**2) | |
|
|
531 | zenith_arr = numpy.arccos(dir_cosw) | |
|
|
532 | azimuth_arr = numpy.arctan2(theta_x,theta_y) + azimuth*math.pi/180 | |
|
|
653 | 533 | |
|
|
654 | #Dealias | |
|
|
655 | phaseArrival = numpy.angle(numpy.exp(1j*phaseArrival)) | |
|
|
656 | # indAlias = numpy.where(phaseArrival > numpy.pi) | |
|
|
657 | # phaseArrival[indAlias] -= 2*numpy.pi | |
|
|
658 | # indAlias = numpy.where(phaseArrival < -numpy.pi) | |
|
|
659 | # phaseArrival[indAlias] += 2*numpy.pi | |
|
|
534 | dir_cosu = numpy.sin(azimuth_arr)*numpy.sin(zenith_arr) | |
|
|
535 | dir_cosv = numpy.cos(azimuth_arr)*numpy.sin(zenith_arr) | |
|
|
660 | 536 | |
|
|
661 | return phaseDiff, phaseArrival | |
|
|
662 | ||
|
|
663 | def __coherentDetection(self, volts, timeSegment, timeInterval, pairslist, thresh): | |
|
|
664 | #this function will run the coherent detection used in Holdworth et al. 2004 and return the net power | |
|
|
665 | #find the phase shifts of each channel over 1 second intervals | |
|
|
666 | #only look at ranges below the beacon signal | |
|
|
667 | numProfPerBlock = numpy.ceil(timeSegment/timeInterval) | |
|
|
668 | numBlocks = int(volts.shape[1]/numProfPerBlock) | |
|
|
669 | numHeights = volts.shape[2] | |
|
|
670 | nChannel = volts.shape[0] | |
|
|
671 | voltsCohDet = volts.copy() | |
|
|
537 | return azimuth_arr, zenith_arr, dir_cosu, dir_cosv, dir_cosw | |
|
|
538 | ||
|
|
539 | def __calculateMatA(self, dir_cosu, dir_cosv, dir_cosw, horOnly): | |
|
|
672 | 540 | |
|
|
673 | pairsarray = numpy.array(pairslist) | |
|
|
674 | indSides = pairsarray[:,1] | |
|
|
675 | # indSides = numpy.array(range(nChannel)) | |
|
|
676 | # indSides = numpy.delete(indSides, indCenter) | |
|
|
677 | 541 | # |
|
|
678 | # listCenter = numpy.array_split(volts[indCenter,:,:], numBlocks, 0) | |
|
|
679 | listBlocks = numpy.array_split(volts, numBlocks, 1) | |
|
|
542 | if horOnly: | |
|
|
543 | A = numpy.c_[dir_cosu,dir_cosv] | |
|
|
544 | else: | |
|
|
545 | A = numpy.c_[dir_cosu,dir_cosv,dir_cosw] | |
|
|
546 | A = numpy.asmatrix(A) | |
|
|
547 | A1 = numpy.linalg.inv(A.transpose()*A)*A.transpose() | |
|
|
548 | ||
|
|
549 | return A1 | |
|
|
550 | ||
|
|
551 | def __correctValues(self, heiRang, phi, velRadial, SNR): | |
|
|
552 | listPhi = phi.tolist() | |
|
|
553 | maxid = listPhi.index(max(listPhi)) | |
|
|
554 | minid = listPhi.index(min(listPhi)) | |
|
|
680 | 555 | |
|
|
681 | startInd = 0 | |
|
|
682 | endInd = 0 | |
|
|
556 | rango = range(len(phi)) | |
|
|
557 | # rango = numpy.delete(rango,maxid) | |
|
|
558 | ||
|
|
559 | heiRang1 = heiRang*math.cos(phi[maxid]) | |
|
|
560 | heiRangAux = heiRang*math.cos(phi[minid]) | |
|
|
561 | indOut = (heiRang1 < heiRangAux[0]).nonzero() | |
|
|
562 | heiRang1 = numpy.delete(heiRang1,indOut) | |
|
|
563 | ||
|
|
564 | velRadial1 = numpy.zeros([len(phi),len(heiRang1)]) | |
|
|
565 | SNR1 = numpy.zeros([len(phi),len(heiRang1)]) | |
|
|
566 | ||
|
|
567 | for i in rango: | |
|
|
568 | x = heiRang*math.cos(phi[i]) | |
|
|
569 | y1 = velRadial[i,:] | |
|
|
570 | f1 = interpolate.interp1d(x,y1,kind = 'cubic') | |
|
|
683 | 571 | |
|
|
684 | for i in range(numBlocks): | |
|
|
685 | startInd = endInd | |
|
|
686 | endInd = endInd + listBlocks[i].shape[1] | |
|
|
572 | x1 = heiRang1 | |
|
|
573 | y11 = f1(x1) | |
|
|
687 | 574 | |
|
|
688 |
|
|
|
|
689 | # arrayBlockCenter = listCenter[i] | |
|
|
575 | y2 = SNR[i,:] | |
|
|
576 | f2 = interpolate.interp1d(x,y2,kind = 'cubic') | |
|
|
577 | y21 = f2(x1) | |
|
|
690 | 578 | |
|
|
691 | #Estimate the Phase Difference | |
|
|
692 | phaseDiff, aux = self.__estimatePhaseDifference(arrayBlock, pairslist) | |
|
|
693 | #Phase Difference RMS | |
|
|
694 | arrayPhaseRMS = numpy.abs(phaseDiff) | |
|
|
695 | phaseRMSaux = numpy.sum(arrayPhaseRMS < thresh,0) | |
|
|
696 | indPhase = numpy.where(phaseRMSaux==4) | |
|
|
697 | #Shifting | |
|
|
698 | if indPhase[0].shape[0] > 0: | |
|
|
699 | for j in range(indSides.size): | |
|
|
700 | arrayBlock[indSides[j],:,indPhase] = self.__shiftPhase(arrayBlock[indSides[j],:,indPhase], phaseDiff[j,indPhase].transpose()) | |
|
|
701 | voltsCohDet[:,startInd:endInd,:] = arrayBlock | |
|
|
702 | ||
|
|
703 | return voltsCohDet | |
|
|
704 | ||
|
|
705 | def __calculateCCF(self, volts, pairslist ,laglist): | |
|
|
579 | velRadial1[i,:] = y11 | |
|
|
580 | SNR1[i,:] = y21 | |
|
|
581 | ||
|
|
582 | return heiRang1, velRadial1, SNR1 | |
|
|
583 | ||
|
|
584 | def __calculateVelUVW(self, A, velRadial): | |
|
|
706 | 585 | |
|
|
707 | nHeights = volts.shape[2] | |
|
|
708 | nPoints = volts.shape[1] | |
|
|
709 | voltsCCF = numpy.zeros((len(pairslist), len(laglist), nHeights),dtype = 'complex') | |
|
|
586 | #Operacion Matricial | |
|
|
587 | # velUVW = numpy.zeros((velRadial.shape[1],3)) | |
|
|
588 | # for ind in range(velRadial.shape[1]): | |
|
|
589 | # velUVW[ind,:] = numpy.dot(A,velRadial[:,ind]) | |
|
|
590 | # velUVW = velUVW.transpose() | |
|
|
591 | velUVW = numpy.zeros((A.shape[0],velRadial.shape[1])) | |
|
|
592 | velUVW[:,:] = numpy.dot(A,velRadial) | |
|
|
710 | 593 | |
|
|
711 | for i in range(len(pairslist)): | |
|
|
712 | volts1 = volts[pairslist[i][0]] | |
|
|
713 | volts2 = volts[pairslist[i][1]] | |
|
|
714 | ||
|
|
715 | for t in range(len(laglist)): | |
|
|
716 | idxT = laglist[t] | |
|
|
717 | if idxT >= 0: | |
|
|
718 | vStacked = numpy.vstack((volts2[idxT:,:], | |
|
|
719 | numpy.zeros((idxT, nHeights),dtype='complex'))) | |
|
|
720 | else: | |
|
|
721 | vStacked = numpy.vstack((numpy.zeros((-idxT, nHeights),dtype='complex'), | |
|
|
722 | volts2[:(nPoints + idxT),:])) | |
|
|
723 | voltsCCF[i,t,:] = numpy.sum((numpy.conjugate(volts1)*vStacked),axis=0) | |
|
|
594 | ||
|
|
595 | return velUVW | |
|
|
724 | 596 | |
|
|
725 | vStacked = None | |
|
|
726 | return voltsCCF | |
|
|
597 | # def techniqueDBS(self, velRadial0, dirCosx, disrCosy, azimuth, correct, horizontalOnly, heiRang, SNR0): | |
|
|
598 | def techniqueDBS(self, velRadial0, heiRang, SNR0, kwargs): | |
|
|
599 | """ | |
|
|
600 | Function that implements Doppler Beam Swinging (DBS) technique. | |
|
|
727 | 601 |
|
|
|
728 | def __getNoise(self, power, timeSegment, timeInterval): | |
|
|
729 | numProfPerBlock = numpy.ceil(timeSegment/timeInterval) | |
|
|
730 | numBlocks = int(power.shape[0]/numProfPerBlock) | |
|
|
731 | numHeights = power.shape[1] | |
|
|
732 | ||
|
|
733 | listPower = numpy.array_split(power, numBlocks, 0) | |
|
|
734 | noise = numpy.zeros((power.shape[0], power.shape[1])) | |
|
|
735 | noise1 = numpy.zeros((power.shape[0], power.shape[1])) | |
|
|
602 | Input: Radial velocities, Direction cosines (x and y) of the Beam, Antenna azimuth, | |
|
|
603 | Direction correction (if necessary), Ranges and SNR | |
|
|
736 | 604 |
|
|
|
737 | startInd = 0 | |
|
|
738 | endInd = 0 | |
|
|
605 | Output: Winds estimation (Zonal, Meridional and Vertical) | |
|
|
739 | 606 |
|
|
|
740 | for i in range(numBlocks): #split por canal | |
|
|
741 | startInd = endInd | |
|
|
742 | endInd = endInd + listPower[i].shape[0] | |
|
|
743 | ||
|
|
744 | arrayBlock = listPower[i] | |
|
|
745 | noiseAux = numpy.mean(arrayBlock, 0) | |
|
|
746 | # noiseAux = numpy.median(noiseAux) | |
|
|
747 | # noiseAux = numpy.mean(arrayBlock) | |
|
|
748 | noise[startInd:endInd,:] = noise[startInd:endInd,:] + noiseAux | |
|
|
749 | ||
|
|
750 | noiseAux1 = numpy.mean(arrayBlock) | |
|
|
751 | noise1[startInd:endInd,:] = noise1[startInd:endInd,:] + noiseAux1 | |
|
|
752 | ||
|
|
753 | return noise, noise1 | |
|
|
754 | ||
|
|
755 | def __findMeteors(self, power, thresh): | |
|
|
756 | nProf = power.shape[0] | |
|
|
757 | nHeights = power.shape[1] | |
|
|
758 | listMeteors = [] | |
|
|
607 | Parameters affected: Winds, height range, SNR | |
|
|
608 | """ | |
|
|
759 | 609 | |
|
|
760 | for i in range(nHeights): | |
|
|
761 | powerAux = power[:,i] | |
|
|
762 | threshAux = thresh[:,i] | |
|
|
763 |
|
|
|
|
764 | indUPthresh = numpy.where(powerAux > threshAux)[0] | |
|
|
765 | indDNthresh = numpy.where(powerAux <= threshAux)[0] | |
|
|
766 | ||
|
|
767 | j = 0 | |
|
|
768 | ||
|
|
769 | while (j < indUPthresh.size - 2): | |
|
|
770 | if (indUPthresh[j + 2] == indUPthresh[j] + 2): | |
|
|
771 | indDNAux = numpy.where(indDNthresh > indUPthresh[j]) | |
|
|
772 | indDNthresh = indDNthresh[indDNAux] | |
|
|
773 | ||
|
|
774 | if (indDNthresh.size > 0): | |
|
|
775 | indEnd = indDNthresh[0] - 1 | |
|
|
776 | indInit = indUPthresh[j] | |
|
|
777 | ||
|
|
778 | meteor = powerAux[indInit:indEnd + 1] | |
|
|
779 | indPeak = meteor.argmax() + indInit | |
|
|
780 | FLA = sum(numpy.conj(meteor)*numpy.hstack((meteor[1:],0))) | |
|
|
781 | ||
|
|
782 | listMeteors.append(numpy.array([i,indInit,indPeak,indEnd,FLA])) #CHEQUEAR!!!!! | |
|
|
783 | j = numpy.where(indUPthresh == indEnd)[0] + 1 | |
|
|
784 | else: j+=1 | |
|
|
785 | else: j+=1 | |
|
|
786 |
|
|
|
|
787 | return listMeteors | |
|
|
610 | if kwargs.has_key('dirCosx') and kwargs.has_key('dirCosy'): | |
|
|
611 | theta_x = numpy.array(kwargs['dirCosx']) | |
|
|
612 | theta_y = numpy.array(kwargs['dirCosy']) | |
|
|
613 | else: | |
|
|
614 | elev = numpy.array(kwargs['elevation']) | |
|
|
615 | azim = numpy.array(kwargs['azimuth']) | |
|
|
616 | theta_x, theta_y = self.__calculateCosDir(elev, azim) | |
|
|
617 | azimuth = kwargs['correctAzimuth'] | |
|
|
618 | if kwargs.has_key('horizontalOnly'): | |
|
|
619 | horizontalOnly = kwargs['horizontalOnly'] | |
|
|
620 | else: horizontalOnly = False | |
|
|
621 | if kwargs.has_key('correctFactor'): | |
|
|
622 | correctFactor = kwargs['correctFactor'] | |
|
|
623 | else: correctFactor = 1 | |
|
|
624 | if kwargs.has_key('channelList'): | |
|
|
625 | channelList = kwargs['channelList'] | |
|
|
626 | if len(channelList) == 2: | |
|
|
627 | horizontalOnly = True | |
|
|
628 | arrayChannel = numpy.array(channelList) | |
|
|
629 | param = param[arrayChannel,:,:] | |
|
|
630 | theta_x = theta_x[arrayChannel] | |
|
|
631 | theta_y = theta_y[arrayChannel] | |
|
|
632 | ||
|
|
633 | azimuth_arr, zenith_arr, dir_cosu, dir_cosv, dir_cosw = self.__calculateAngles(theta_x, theta_y, azimuth) | |
|
|
634 | heiRang1, velRadial1, SNR1 = self.__correctValues(heiRang, zenith_arr, correctFactor*velRadial0, SNR0) | |
|
|
635 | A = self.__calculateMatA(dir_cosu, dir_cosv, dir_cosw, horizontalOnly) | |
|
|
636 | ||
|
|
637 | #Calculo de Componentes de la velocidad con DBS | |
|
|
638 | winds = self.__calculateVelUVW(A,velRadial1) | |
|
|
639 | ||
|
|
640 | return winds, heiRang1, SNR1 | |
|
|
788 | 641 | |
|
|
789 | def __removeMultipleDetections(self,listMeteors, rangeLimit, timeLimit): | |
|
|
642 | def __calculateDistance(self, posx, posy, pairsCrossCorr, pairsList, pairs, azimuth = None): | |
|
|
790 | 643 | |
|
|
791 |
|
|
|
|
792 | listMeteors1 = [] | |
|
|
644 | posx = numpy.asarray(posx) | |
|
|
645 | posy = numpy.asarray(posy) | |
|
|
793 | 646 | |
|
|
794 | while arrayMeteors.shape[0] > 0: | |
|
|
795 | FLAs = arrayMeteors[:,4] | |
|
|
796 | maxFLA = FLAs.argmax() | |
|
|
797 | listMeteors1.append(arrayMeteors[maxFLA,:]) | |
|
|
798 | ||
|
|
799 | MeteorInitTime = arrayMeteors[maxFLA,1] | |
|
|
800 | MeteorEndTime = arrayMeteors[maxFLA,3] | |
|
|
801 | MeteorHeight = arrayMeteors[maxFLA,0] | |
|
|
802 |
|
|
|
|
803 | #Check neighborhood | |
|
|
804 | maxHeightIndex = MeteorHeight + rangeLimit | |
|
|
805 | minHeightIndex = MeteorHeight - rangeLimit | |
|
|
806 | minTimeIndex = MeteorInitTime - timeLimit | |
|
|
807 | maxTimeIndex = MeteorEndTime + timeLimit | |
|
|
808 |
|
|
|
|
809 | #Check Heights | |
|
|
810 | indHeight = numpy.logical_and(arrayMeteors[:,0] >= minHeightIndex, arrayMeteors[:,0] <= maxHeightIndex) | |
|
|
811 | indTime = numpy.logical_and(arrayMeteors[:,3] >= minTimeIndex, arrayMeteors[:,1] <= maxTimeIndex) | |
|
|
812 | indBoth = numpy.where(numpy.logical_and(indTime,indHeight)) | |
|
|
647 | #Rotacion Inversa para alinear con el azimuth | |
|
|
648 | if azimuth!= None: | |
|
|
649 | azimuth = azimuth*math.pi/180 | |
|
|
650 | posx1 = posx*math.cos(azimuth) + posy*math.sin(azimuth) | |
|
|
651 | posy1 = -posx*math.sin(azimuth) + posy*math.cos(azimuth) | |
|
|
652 | else: | |
|
|
653 | posx1 = posx | |
|
|
654 | posy1 = posy | |
|
|
655 | ||
|
|
656 | #Calculo de Distancias | |
|
|
657 | distx = numpy.zeros(pairsCrossCorr.size) | |
|
|
658 | disty = numpy.zeros(pairsCrossCorr.size) | |
|
|
659 | dist = numpy.zeros(pairsCrossCorr.size) | |
|
|
660 | ang = numpy.zeros(pairsCrossCorr.size) | |
|
|
661 | ||
|
|
662 | for i in range(pairsCrossCorr.size): | |
|
|
663 | distx[i] = posx1[pairsList[pairsCrossCorr[i]][1]] - posx1[pairsList[pairsCrossCorr[i]][0]] | |
|
|
664 | disty[i] = posy1[pairsList[pairsCrossCorr[i]][1]] - posy1[pairsList[pairsCrossCorr[i]][0]] | |
|
|
665 | dist[i] = numpy.sqrt(distx[i]**2 + disty[i]**2) | |
|
|
666 | ang[i] = numpy.arctan2(disty[i],distx[i]) | |
|
|
667 | #Calculo de Matrices | |
|
|
668 | nPairs = len(pairs) | |
|
|
669 | ang1 = numpy.zeros((nPairs, 2, 1)) | |
|
|
670 | dist1 = numpy.zeros((nPairs, 2, 1)) | |
|
|
671 | ||
|
|
672 | for j in range(nPairs): | |
|
|
673 | dist1[j,0,0] = dist[pairs[j][0]] | |
|
|
674 | dist1[j,1,0] = dist[pairs[j][1]] | |
|
|
675 | ang1[j,0,0] = ang[pairs[j][0]] | |
|
|
676 | ang1[j,1,0] = ang[pairs[j][1]] | |
|
|
813 | 677 | |
|
|
814 | arrayMeteors = numpy.delete(arrayMeteors, indBoth, axis = 0) | |
|
|
678 | return distx,disty, dist1,ang1 | |
|
|
679 | ||
|
|
680 | def __calculateVelVer(self, phase, lagTRange, _lambda): | |
|
|
681 | ||
|
|
682 | Ts = lagTRange[1] - lagTRange[0] | |
|
|
683 | velW = -_lambda*phase/(4*math.pi*Ts) | |
|
|
815 | 684 | |
|
|
816 |
return |
|
|
|
685 | return velW | |
|
|
817 | 686 | |
|
|
818 | def __meteorReestimation(self, listMeteors, volts, pairslist, thresh, noise, timeInterval,frequency): | |
|
|
819 |
n |
|
|
|
820 | nChannel = volts.shape[0] | |
|
|
687 | def __calculateVelHorDir(self, dist, tau1, tau2, ang): | |
|
|
688 | nPairs = tau1.shape[0] | |
|
|
689 | vel = numpy.zeros((nPairs,3,tau1.shape[2])) | |
|
|
821 | 690 | |
|
|
822 | thresholdPhase = thresh[0] | |
|
|
823 | thresholdNoise = thresh[1] | |
|
|
824 | thresholdDB = float(thresh[2]) | |
|
|
691 | angCos = numpy.cos(ang) | |
|
|
692 | angSin = numpy.sin(ang) | |
|
|
825 | 693 | |
|
|
826 | thresholdDB1 = 10**(thresholdDB/10) | |
|
|
827 | pairsarray = numpy.array(pairslist) | |
|
|
828 | indSides = pairsarray[:,1] | |
|
|
694 | vel0 = dist*tau1/(2*tau2**2) | |
|
|
695 | vel[:,0,:] = (vel0*angCos).sum(axis = 1) | |
|
|
696 | vel[:,1,:] = (vel0*angSin).sum(axis = 1) | |
|
|
829 | 697 | |
|
|
830 | pairslist1 = list(pairslist) | |
|
|
831 | pairslist1.append((0,1)) | |
|
|
832 | pairslist1.append((3,4)) | |
|
|
698 | ind = numpy.where(numpy.isinf(vel)) | |
|
|
699 | vel[ind] = numpy.nan | |
|
|
700 | ||
|
|
701 | return vel | |
|
|
702 | ||
|
|
703 | def __getPairsAutoCorr(self, pairsList, nChannels): | |
|
|
833 | 704 | |
|
|
834 | listMeteors1 = [] | |
|
|
835 | listPowerSeries = [] | |
|
|
836 | listVoltageSeries = [] | |
|
|
837 | #volts has the war data | |
|
|
705 | pairsAutoCorr = numpy.zeros(nChannels, dtype = 'int')*numpy.nan | |
|
|
706 | ||
|
|
707 | for l in range(len(pairsList)): | |
|
|
708 | firstChannel = pairsList[l][0] | |
|
|
709 | secondChannel = pairsList[l][1] | |
|
|
710 | ||
|
|
711 | #Obteniendo pares de Autocorrelacion | |
|
|
712 | if firstChannel == secondChannel: | |
|
|
713 | pairsAutoCorr[firstChannel] = int(l) | |
|
|
714 | ||
|
|
715 | pairsAutoCorr = pairsAutoCorr.astype(int) | |
|
|
838 | 716 | |
|
|
839 | if frequency == 30e6: | |
|
|
840 | timeLag = 45*10**-3 | |
|
|
841 | else: | |
|
|
842 | timeLag = 15*10**-3 | |
|
|
843 | lag = numpy.ceil(timeLag/timeInterval) | |
|
|
717 | pairsCrossCorr = range(len(pairsList)) | |
|
|
718 | pairsCrossCorr = numpy.delete(pairsCrossCorr,pairsAutoCorr) | |
|
|
844 | 719 | |
|
|
845 | for i in range(len(listMeteors)): | |
|
|
846 | ||
|
|
847 | ###################### 3.6 - 3.7 PARAMETERS REESTIMATION ######################### | |
|
|
848 | meteorAux = numpy.zeros(16) | |
|
|
849 | ||
|
|
850 | #Loading meteor Data (mHeight, mStart, mPeak, mEnd) | |
|
|
851 | mHeight = listMeteors[i][0] | |
|
|
852 | mStart = listMeteors[i][1] | |
|
|
853 | mPeak = listMeteors[i][2] | |
|
|
854 | mEnd = listMeteors[i][3] | |
|
|
855 | ||
|
|
856 | #get the volt data between the start and end times of the meteor | |
|
|
857 | meteorVolts = volts[:,mStart:mEnd+1,mHeight] | |
|
|
858 | meteorVolts = meteorVolts.reshape(meteorVolts.shape[0], meteorVolts.shape[1], 1) | |
|
|
859 | ||
|
|
860 | #3.6. Phase Difference estimation | |
|
|
861 | phaseDiff, aux = self.__estimatePhaseDifference(meteorVolts, pairslist) | |
|
|
862 | ||
|
|
863 | #3.7. Phase difference removal & meteor start, peak and end times reestimated | |
|
|
864 | #meteorVolts0.- all Channels, all Profiles | |
|
|
865 | meteorVolts0 = volts[:,:,mHeight] | |
|
|
866 | meteorThresh = noise[:,mHeight]*thresholdNoise | |
|
|
867 | meteorNoise = noise[:,mHeight] | |
|
|
868 | meteorVolts0[indSides,:] = self.__shiftPhase(meteorVolts0[indSides,:], phaseDiff) #Phase Shifting | |
|
|
869 | powerNet0 = numpy.nansum(numpy.abs(meteorVolts0)**2, axis = 0) #Power | |
|
|
870 | ||
|
|
871 | #Times reestimation | |
|
|
872 | mStart1 = numpy.where(powerNet0[:mPeak] < meteorThresh[:mPeak])[0] | |
|
|
873 | if mStart1.size > 0: | |
|
|
874 | mStart1 = mStart1[-1] + 1 | |
|
|
875 | ||
|
|
876 | else: | |
|
|
877 | mStart1 = mPeak | |
|
|
878 | ||
|
|
879 | mEnd1 = numpy.where(powerNet0[mPeak:] < meteorThresh[mPeak:])[0][0] + mPeak - 1 | |
|
|
880 | mEndDecayTime1 = numpy.where(powerNet0[mPeak:] < meteorNoise[mPeak:])[0] | |
|
|
881 | if mEndDecayTime1.size == 0: | |
|
|
882 | mEndDecayTime1 = powerNet0.size | |
|
|
883 | else: | |
|
|
884 | mEndDecayTime1 = mEndDecayTime1[0] + mPeak - 1 | |
|
|
885 | # mPeak1 = meteorVolts0[mStart1:mEnd1 + 1].argmax() | |
|
|
886 | ||
|
|
887 | #meteorVolts1.- all Channels, from start to end | |
|
|
888 | meteorVolts1 = meteorVolts0[:,mStart1:mEnd1 + 1] | |
|
|
889 | meteorVolts2 = meteorVolts0[:,mPeak + lag:mEnd1 + 1] | |
|
|
890 | if meteorVolts2.shape[1] == 0: | |
|
|
891 | meteorVolts2 = meteorVolts0[:,mPeak:mEnd1 + 1] | |
|
|
892 | meteorVolts1 = meteorVolts1.reshape(meteorVolts1.shape[0], meteorVolts1.shape[1], 1) | |
|
|
893 | meteorVolts2 = meteorVolts2.reshape(meteorVolts2.shape[0], meteorVolts2.shape[1], 1) | |
|
|
894 | ##################### END PARAMETERS REESTIMATION ######################### | |
|
|
895 | ||
|
|
896 | ##################### 3.8 PHASE DIFFERENCE REESTIMATION ######################## | |
|
|
897 | # if mEnd1 - mStart1 > 4: #Error Number 6: echo less than 5 samples long; too short for analysis | |
|
|
898 | if meteorVolts2.shape[1] > 0: | |
|
|
899 | #Phase Difference re-estimation | |
|
|
900 | phaseDiff1, phaseDiffint = self.__estimatePhaseDifference(meteorVolts2, pairslist1) #Phase Difference Estimation | |
|
|
901 | # phaseDiff1, phaseDiffint = self.estimatePhaseDifference(meteorVolts2, pairslist) | |
|
|
902 | meteorVolts2 = meteorVolts2.reshape(meteorVolts2.shape[0], meteorVolts2.shape[1]) | |
|
|
903 | phaseDiff11 = numpy.reshape(phaseDiff1, (phaseDiff1.shape[0],1)) | |
|
|
904 | meteorVolts2[indSides,:] = self.__shiftPhase(meteorVolts2[indSides,:], phaseDiff11[0:4]) #Phase Shifting | |
|
|
905 | ||
|
|
906 | #Phase Difference RMS | |
|
|
907 | phaseRMS1 = numpy.sqrt(numpy.mean(numpy.square(phaseDiff1))) | |
|
|
908 | powerNet1 = numpy.nansum(numpy.abs(meteorVolts1[:,:])**2,0) | |
|
|
909 | #Data from Meteor | |
|
|
910 | mPeak1 = powerNet1.argmax() + mStart1 | |
|
|
911 | mPeakPower1 = powerNet1.max() | |
|
|
912 | noiseAux = sum(noise[mStart1:mEnd1 + 1,mHeight]) | |
|
|
913 | mSNR1 = (sum(powerNet1)-noiseAux)/noiseAux | |
|
|
914 | Meteor1 = numpy.array([mHeight, mStart1, mPeak1, mEnd1, mPeakPower1, mSNR1, phaseRMS1]) | |
|
|
915 | Meteor1 = numpy.hstack((Meteor1,phaseDiffint)) | |
|
|
916 | PowerSeries = powerNet0[mStart1:mEndDecayTime1 + 1] | |
|
|
917 | #Vectorize | |
|
|
918 | meteorAux[0:7] = [mHeight, mStart1, mPeak1, mEnd1, mPeakPower1, mSNR1, phaseRMS1] | |
|
|
919 | meteorAux[7:11] = phaseDiffint[0:4] | |
|
|
920 | ||
|
|
921 | #Rejection Criterions | |
|
|
922 | if phaseRMS1 > thresholdPhase: #Error Number 17: Phase variation | |
|
|
923 | meteorAux[-1] = 17 | |
|
|
924 | elif mSNR1 < thresholdDB1: #Error Number 1: SNR < threshold dB | |
|
|
925 | meteorAux[-1] = 1 | |
|
|
926 | ||
|
|
927 | ||
|
|
928 | else: | |
|
|
929 | meteorAux[0:4] = [mHeight, mStart, mPeak, mEnd] | |
|
|
930 | meteorAux[-1] = 6 #Error Number 6: echo less than 5 samples long; too short for analysis | |
|
|
931 | PowerSeries = 0 | |
|
|
932 | ||
|
|
933 | listMeteors1.append(meteorAux) | |
|
|
934 | listPowerSeries.append(PowerSeries) | |
|
|
935 | listVoltageSeries.append(meteorVolts1) | |
|
|
936 | ||
|
|
937 | return listMeteors1, listPowerSeries, listVoltageSeries | |
|
|
938 | ||
|
|
939 | def __estimateDecayTime(self, listMeteors, listPower, timeInterval, frequency): | |
|
|
720 | return pairsAutoCorr, pairsCrossCorr | |
|
|
721 | ||
|
|
722 | def techniqueSA(self, pairsSelected, pairsList, nChannels, tau, azimuth, _lambda, position_x, position_y, lagTRange, correctFactor): | |
|
|
723 | """ | |
|
|
724 | Function that implements Spaced Antenna (SA) technique. | |
|
|
940 | 725 |
|
|
|
941 | threshError = 10 | |
|
|
942 | #Depending if it is 30 or 50 MHz | |
|
|
943 | if frequency == 30e6: | |
|
|
944 | timeLag = 45*10**-3 | |
|
|
945 | else: | |
|
|
946 | timeLag = 15*10**-3 | |
|
|
947 | lag = numpy.ceil(timeLag/timeInterval) | |
|
|
726 | Input: Radial velocities, Direction cosines (x and y) of the Beam, Antenna azimuth, | |
|
|
727 | Direction correction (if necessary), Ranges and SNR | |
|
|
948 | 728 |
|
|
|
949 | listMeteors1 = [] | |
|
|
729 | Output: Winds estimation (Zonal, Meridional and Vertical) | |
|
|
950 | 730 |
|
|
|
951 | for i in range(len(listMeteors)): | |
|
|
952 | meteorPower = listPower[i] | |
|
|
953 | meteorAux = listMeteors[i] | |
|
|
954 | ||
|
|
955 | if meteorAux[-1] == 0: | |
|
|
956 | ||
|
|
957 | try: | |
|
|
958 | indmax = meteorPower.argmax() | |
|
|
959 | indlag = indmax + lag | |
|
|
960 | ||
|
|
961 | y = meteorPower[indlag:] | |
|
|
962 | x = numpy.arange(0, y.size)*timeLag | |
|
|
963 | ||
|
|
964 | #first guess | |
|
|
965 | a = y[0] | |
|
|
966 | tau = timeLag | |
|
|
967 | #exponential fit | |
|
|
968 | popt, pcov = optimize.curve_fit(self.__exponential_function, x, y, p0 = [a, tau]) | |
|
|
969 | y1 = self.__exponential_function(x, *popt) | |
|
|
970 | #error estimation | |
|
|
971 | error = sum((y - y1)**2)/(numpy.var(y)*(y.size - popt.size)) | |
|
|
972 | ||
|
|
973 | decayTime = popt[1] | |
|
|
974 | riseTime = indmax*timeInterval | |
|
|
975 | meteorAux[11:13] = [decayTime, error] | |
|
|
976 | ||
|
|
977 | #Table items 7, 8 and 11 | |
|
|
978 | if (riseTime > 0.3): #Number 7: Echo rise exceeds 0.3s | |
|
|
979 | meteorAux[-1] = 7 | |
|
|
980 | elif (decayTime < 2*riseTime) : #Number 8: Echo decay time less than than twice rise time | |
|
|
981 | meteorAux[-1] = 8 | |
|
|
982 | if (error > threshError): #Number 11: Poor fit to amplitude for estimation of decay time | |
|
|
983 | meteorAux[-1] = 11 | |
|
|
984 | ||
|
|
985 | ||
|
|
986 | except: | |
|
|
987 | meteorAux[-1] = 11 | |
|
|
988 | ||
|
|
989 | ||
|
|
990 | listMeteors1.append(meteorAux) | |
|
|
731 | Parameters affected: Winds | |
|
|
732 | """ | |
|
|
733 | #Cross Correlation pairs obtained | |
|
|
734 | pairsAutoCorr, pairsCrossCorr = self.__getPairsAutoCorr(pairsList, nChannels) | |
|
|
735 | pairsArray = numpy.array(pairsList)[pairsCrossCorr] | |
|
|
736 | pairsSelArray = numpy.array(pairsSelected) | |
|
|
737 | pairs = [] | |
|
|
991 | 738 | |
|
|
992 | return listMeteors1 | |
|
|
993 | ||
|
|
994 | #Exponential Function | |
|
|
739 | #Wind estimation pairs obtained | |
|
|
740 | for i in range(pairsSelArray.shape[0]/2): | |
|
|
741 | ind1 = numpy.where(numpy.all(pairsArray == pairsSelArray[2*i], axis = 1))[0][0] | |
|
|
742 | ind2 = numpy.where(numpy.all(pairsArray == pairsSelArray[2*i + 1], axis = 1))[0][0] | |
|
|
743 | pairs.append((ind1,ind2)) | |
|
|
744 | ||
|
|
745 | indtau = tau.shape[0]/2 | |
|
|
746 | tau1 = tau[:indtau,:] | |
|
|
747 | tau2 = tau[indtau:-1,:] | |
|
|
748 | tau1 = tau1[pairs,:] | |
|
|
749 | tau2 = tau2[pairs,:] | |
|
|
750 | phase1 = tau[-1,:] | |
|
|
751 | ||
|
|
752 | #--------------------------------------------------------------------- | |
|
|
753 | #Metodo Directo | |
|
|
754 | distx, disty, dist, ang = self.__calculateDistance(position_x, position_y, pairsCrossCorr, pairsList, pairs,azimuth) | |
|
|
755 | winds = self.__calculateVelHorDir(dist, tau1, tau2, ang) | |
|
|
756 | winds = stats.nanmean(winds, axis=0) | |
|
|
757 | #--------------------------------------------------------------------- | |
|
|
758 | #Metodo General | |
|
|
759 | # distx, disty, dist = self.calculateDistance(position_x,position_y,pairsCrossCorr, pairsList, azimuth) | |
|
|
760 | # #Calculo Coeficientes de Funcion de Correlacion | |
|
|
761 | # F,G,A,B,H = self.calculateCoef(tau1,tau2,distx,disty,n) | |
|
|
762 | # #Calculo de Velocidades | |
|
|
763 | # winds = self.calculateVelUV(F,G,A,B,H) | |
|
|
995 | 764 | |
|
|
996 | def __exponential_function(self, x, a, tau): | |
|
|
997 | y = a*numpy.exp(-x/tau) | |
|
|
998 | return y | |
|
|
765 | #--------------------------------------------------------------------- | |
|
|
766 | winds[2,:] = self.__calculateVelVer(phase1, lagTRange, _lambda) | |
|
|
767 | winds = correctFactor*winds | |
|
|
768 | return winds | |
|
|
999 | 769 | |
|
|
1000 | def __getRadialVelocity(self, listMeteors, listVolts, radialStdThresh, pairslist, timeInterval): | |
|
|
770 | def __checkTime(self, currentTime, paramInterval, outputInterval): | |
|
|
1001 | 771 | |
|
|
1002 | pairslist1 = list(pairslist) | |
|
|
1003 | pairslist1.append((0,1)) | |
|
|
1004 | pairslist1.append((3,4)) | |
|
|
1005 | numPairs = len(pairslist1) | |
|
|
1006 | #Time Lag | |
|
|
1007 | timeLag = 45*10**-3 | |
|
|
1008 | c = 3e8 | |
|
|
1009 | lag = numpy.ceil(timeLag/timeInterval) | |
|
|
1010 | freq = 30e6 | |
|
|
772 | dataTime = currentTime + paramInterval | |
|
|
773 | deltaTime = dataTime - self.__initime | |
|
|
1011 | 774 | |
|
|
1012 | listMeteors1 = [] | |
|
|
775 | if deltaTime >= outputInterval or deltaTime < 0: | |
|
|
776 | self.__dataReady = True | |
|
|
777 | return | |
|
|
778 | ||
|
|
779 | def techniqueMeteors(self, arrayMeteor, meteorThresh, heightMin, heightMax): | |
|
|
780 | ''' | |
|
|
781 | Function that implements winds estimation technique with detected meteors. | |
|
|
1013 | 782 |
|
|
|
1014 | for i in range(len(listMeteors)): | |
|
|
1015 | meteorAux = listMeteors[i] | |
|
|
1016 | if meteorAux[-1] == 0: | |
|
|
1017 | mStart = listMeteors[i][1] | |
|
|
1018 | mPeak = listMeteors[i][2] | |
|
|
1019 | mLag = mPeak - mStart + lag | |
|
|
1020 | ||
|
|
1021 | #get the volt data between the start and end times of the meteor | |
|
|
1022 | meteorVolts = listVolts[i] | |
|
|
1023 | meteorVolts = meteorVolts.reshape(meteorVolts.shape[0], meteorVolts.shape[1], 1) | |
|
|
1024 | ||
|
|
1025 | #Get CCF | |
|
|
1026 | allCCFs = self.__calculateCCF(meteorVolts, pairslist1, [-2,-1,0,1,2]) | |
|
|
1027 |
|
|
|
|
1028 | #Method 2 | |
|
|
1029 | slopes = numpy.zeros(numPairs) | |
|
|
1030 | time = numpy.array([-2,-1,1,2])*timeInterval | |
|
|
1031 | angAllCCF = numpy.angle(allCCFs[:,[0,1,3,4],0]) | |
|
|
1032 | ||
|
|
1033 | #Correct phases | |
|
|
1034 | derPhaseCCF = angAllCCF[:,1:] - angAllCCF[:,0:-1] | |
|
|
1035 | indDer = numpy.where(numpy.abs(derPhaseCCF) > numpy.pi) | |
|
|
783 | Input: Detected meteors, Minimum meteor quantity to wind estimation | |
|
|
784 | ||
|
|
785 | Output: Winds estimation (Zonal and Meridional) | |
|
|
786 | ||
|
|
787 | Parameters affected: Winds | |
|
|
788 | ''' | |
|
|
789 | # print arrayMeteor.shape | |
|
|
790 | #Settings | |
|
|
791 | nInt = (heightMax - heightMin)/2 | |
|
|
792 | # print nInt | |
|
|
793 | nInt = int(nInt) | |
|
|
794 | # print nInt | |
|
|
795 | winds = numpy.zeros((2,nInt))*numpy.nan | |
|
|
796 | ||
|
|
797 | #Filter errors | |
|
|
798 | error = numpy.where(arrayMeteor[:,-1] == 0)[0] | |
|
|
799 | finalMeteor = arrayMeteor[error,:] | |
|
|
800 | ||
|
|
801 | #Meteor Histogram | |
|
|
802 | finalHeights = finalMeteor[:,2] | |
|
|
803 | hist = numpy.histogram(finalHeights, bins = nInt, range = (heightMin,heightMax)) | |
|
|
804 | nMeteorsPerI = hist[0] | |
|
|
805 | heightPerI = hist[1] | |
|
|
806 | ||
|
|
807 | #Sort of meteors | |
|
|
808 | indSort = finalHeights.argsort() | |
|
|
809 | finalMeteor2 = finalMeteor[indSort,:] | |
|
|
810 | ||
|
|
811 | # Calculating winds | |
|
|
812 | ind1 = 0 | |
|
|
813 | ind2 = 0 | |
|
|
814 | ||
|
|
815 | for i in range(nInt): | |
|
|
816 | nMet = nMeteorsPerI[i] | |
|
|
817 | ind1 = ind2 | |
|
|
818 | ind2 = ind1 + nMet | |
|
|
819 | ||
|
|
820 | meteorAux = finalMeteor2[ind1:ind2,:] | |
|
|
821 | ||
|
|
822 | if meteorAux.shape[0] >= meteorThresh: | |
|
|
823 | vel = meteorAux[:, 6] | |
|
|
824 | zen = meteorAux[:, 4]*numpy.pi/180 | |
|
|
825 | azim = meteorAux[:, 3]*numpy.pi/180 | |
|
|
1036 | 826 | |
|
|
1037 | if indDer[0].shape[0] > 0: | |
|
|
1038 | for i in range(indDer[0].shape[0]): | |
|
|
1039 | signo = -numpy.sign(derPhaseCCF[indDer[0][i],indDer[1][i]]) | |
|
|
1040 | angAllCCF[indDer[0][i],indDer[1][i]+1:] += signo*2*numpy.pi | |
|
|
1041 | ||
|
|
1042 | # fit = scipy.stats.linregress(numpy.array([-2,-1,1,2])*timeInterval, numpy.array([phaseLagN2s[i],phaseLagN1s[i],phaseLag1s[i],phaseLag2s[i]])) | |
|
|
1043 | for j in range(numPairs): | |
|
|
1044 | fit = stats.linregress(time, angAllCCF[j,:]) | |
|
|
1045 | slopes[j] = fit[0] | |
|
|
827 | n = numpy.cos(zen) | |
|
|
828 | # m = (1 - n**2)/(1 - numpy.tan(azim)**2) | |
|
|
829 | # l = m*numpy.tan(azim) | |
|
|
830 | l = numpy.sin(zen)*numpy.sin(azim) | |
|
|
831 | m = numpy.sin(zen)*numpy.cos(azim) | |
|
|
1046 | 832 | |
|
|
1047 | #Remove Outlier | |
|
|
1048 | # indOut = numpy.argmax(numpy.abs(slopes - numpy.mean(slopes))) | |
|
|
1049 |
|
|
|
|
1050 | # indOut = numpy.argmax(numpy.abs(slopes - numpy.mean(slopes))) | |
|
|
1051 | # slopes = numpy.delete(slopes,indOut) | |
|
|
1052 | ||
|
|
1053 | radialVelocity = -numpy.mean(slopes)*(0.25/numpy.pi)*(c/freq) | |
|
|
1054 | radialError = numpy.std(slopes)*(0.25/numpy.pi)*(c/freq) | |
|
|
1055 | meteorAux[-2] = radialError | |
|
|
1056 | meteorAux[-3] = radialVelocity | |
|
|
833 | A = numpy.vstack((l, m)).transpose() | |
|
|
834 | A1 = numpy.dot(numpy.linalg.inv( numpy.dot(A.transpose(),A) ),A.transpose()) | |
|
|
835 | windsAux = numpy.dot(A1, vel) | |
|
|
1057 | 836 | |
|
|
1058 | #Setting Error | |
|
|
1059 | #Number 15: Radial Drift velocity or projected horizontal velocity exceeds 200 m/s | |
|
|
1060 | if numpy.abs(radialVelocity) > 200: | |
|
|
1061 | meteorAux[-1] = 15 | |
|
|
1062 | #Number 12: Poor fit to CCF variation for estimation of radial drift velocity | |
|
|
1063 | elif radialError > radialStdThresh: | |
|
|
1064 | meteorAux[-1] = 12 | |
|
|
837 | winds[0,i] = windsAux[0] | |
|
|
838 | winds[1,i] = windsAux[1] | |
|
|
1065 | 839 | |
|
|
1066 | listMeteors1.append(meteorAux) | |
|
|
1067 | return listMeteors1 | |
|
|
840 | return winds, heightPerI[:-1] | |
|
|
1068 | 841 | |
|
|
1069 | def __setNewArrays(self, listMeteors, date, heiRang): | |
|
|
1070 | ||
|
|
1071 | #New arrays | |
|
|
1072 | arrayMeteors = numpy.array(listMeteors) | |
|
|
1073 | arrayParameters = numpy.zeros((len(listMeteors), 13)) | |
|
|
842 | def techniqueNSM_SA(self, **kwargs): | |
|
|
843 | metArray = kwargs['metArray'] | |
|
|
844 | heightList = kwargs['heightList'] | |
|
|
845 | timeList = kwargs['timeList'] | |
|
|
1074 | 846 | |
|
|
1075 | #Date inclusion | |
|
|
1076 | # date = re.findall(r'\((.*?)\)', date) | |
|
|
1077 | # date = date[0].split(',') | |
|
|
1078 | # date = map(int, date) | |
|
|
1079 | # | |
|
|
1080 | # if len(date)<6: | |
|
|
1081 | # date.append(0) | |
|
|
1082 | # | |
|
|
1083 | # date = [date[0]*10000 + date[1]*100 + date[2], date[3]*10000 + date[4]*100 + date[5]] | |
|
|
1084 | # arrayDate = numpy.tile(date, (len(listMeteors), 1)) | |
|
|
1085 | arrayDate = numpy.tile(date, (len(listMeteors))) | |
|
|
847 | rx_location = kwargs['rx_location'] | |
|
|
848 | groupList = kwargs['groupList'] | |
|
|
849 | azimuth = kwargs['azimuth'] | |
|
|
850 | dfactor = kwargs['dfactor'] | |
|
|
851 | k = kwargs['k'] | |
|
|
1086 | 852 | |
|
|
1087 | #Meteor array | |
|
|
1088 | # arrayMeteors[:,0] = heiRang[arrayMeteors[:,0].astype(int)] | |
|
|
1089 | # arrayMeteors = numpy.hstack((arrayDate, arrayMeteors)) | |
|
|
853 | azimuth1, dist = self.__calculateAzimuth1(rx_location, groupList, azimuth) | |
|
|
854 | d = dist*dfactor | |
|
|
855 | #Phase calculation | |
|
|
856 | metArray1 = self.__getPhaseSlope(metArray, heightList, timeList) | |
|
|
1090 | 857 | |
|
|
1091 | #Parameters Array | |
|
|
1092 | arrayParameters[:,0] = arrayDate #Date | |
|
|
1093 | arrayParameters[:,1] = heiRang[arrayMeteors[:,0].astype(int)] #Range | |
|
|
1094 | arrayParameters[:,6:8] = arrayMeteors[:,-3:-1] #Radial velocity and its error | |
|
|
1095 | arrayParameters[:,8:12] = arrayMeteors[:,7:11] #Phases | |
|
|
1096 | arrayParameters[:,-1] = arrayMeteors[:,-1] #Error | |
|
|
1097 | ||
|
|
858 | metArray1[:,-2] = metArray1[:,-2]*metArray1[:,2]*1000/(k*d[metArray1[:,1].astype(int)]) #angles into velocities | |
|
|
1098 | 859 | |
|
|
1099 | return arrayParameters | |
|
|
1100 | ||
|
|
1101 | # def __getAOA(self, phases, pairsList, error, AOAthresh, azimuth): | |
|
|
1102 | # | |
|
|
1103 | # arrayAOA = numpy.zeros((phases.shape[0],3)) | |
|
|
1104 | # cosdir0, cosdir = self.__getDirectionCosines(phases, pairsList) | |
|
|
1105 | # | |
|
|
1106 | # arrayAOA[:,:2] = self.__calculateAOA(cosdir, azimuth) | |
|
|
1107 | # cosDirError = numpy.sum(numpy.abs(cosdir0 - cosdir), axis = 1) | |
|
|
1108 | # arrayAOA[:,2] = cosDirError | |
|
|
1109 | # | |
|
|
1110 | # azimuthAngle = arrayAOA[:,0] | |
|
|
1111 | # zenithAngle = arrayAOA[:,1] | |
|
|
1112 | # | |
|
|
1113 | # #Setting Error | |
|
|
1114 | # #Number 3: AOA not fesible | |
|
|
1115 | # indInvalid = numpy.where(numpy.logical_and((numpy.logical_or(numpy.isnan(zenithAngle), numpy.isnan(azimuthAngle))),error == 0))[0] | |
|
|
1116 | # error[indInvalid] = 3 | |
|
|
1117 | # #Number 4: Large difference in AOAs obtained from different antenna baselines | |
|
|
1118 | # indInvalid = numpy.where(numpy.logical_and(cosDirError > AOAthresh,error == 0))[0] | |
|
|
1119 | # error[indInvalid] = 4 | |
|
|
1120 | # return arrayAOA, error | |
|
|
1121 | # | |
|
|
1122 | # def __getDirectionCosines(self, arrayPhase, pairsList): | |
|
|
1123 | # | |
|
|
1124 | # #Initializing some variables | |
|
|
1125 | # ang_aux = numpy.array([-8,-7,-6,-5,-4,-3,-2,-1,0,1,2,3,4,5,6,7,8])*2*numpy.pi | |
|
|
1126 | # ang_aux = ang_aux.reshape(1,ang_aux.size) | |
|
|
1127 | # | |
|
|
1128 | # cosdir = numpy.zeros((arrayPhase.shape[0],2)) | |
|
|
1129 | # cosdir0 = numpy.zeros((arrayPhase.shape[0],2)) | |
|
|
1130 | # | |
|
|
1131 | # | |
|
|
1132 | # for i in range(2): | |
|
|
1133 | # #First Estimation | |
|
|
1134 | # phi0_aux = arrayPhase[:,pairsList[i][0]] + arrayPhase[:,pairsList[i][1]] | |
|
|
1135 | # #Dealias | |
|
|
1136 | # indcsi = numpy.where(phi0_aux > numpy.pi) | |
|
|
1137 | # phi0_aux[indcsi] -= 2*numpy.pi | |
|
|
1138 | # indcsi = numpy.where(phi0_aux < -numpy.pi) | |
|
|
1139 | # phi0_aux[indcsi] += 2*numpy.pi | |
|
|
1140 | # #Direction Cosine 0 | |
|
|
1141 | # cosdir0[:,i] = -(phi0_aux)/(2*numpy.pi*0.5) | |
|
|
1142 | # | |
|
|
1143 | # #Most-Accurate Second Estimation | |
|
|
1144 | # phi1_aux = arrayPhase[:,pairsList[i][0]] - arrayPhase[:,pairsList[i][1]] | |
|
|
1145 | # phi1_aux = phi1_aux.reshape(phi1_aux.size,1) | |
|
|
1146 | # #Direction Cosine 1 | |
|
|
1147 | # cosdir1 = -(phi1_aux + ang_aux)/(2*numpy.pi*4.5) | |
|
|
1148 | # | |
|
|
1149 | # #Searching the correct Direction Cosine | |
|
|
1150 | # cosdir0_aux = cosdir0[:,i] | |
|
|
1151 | # cosdir0_aux = cosdir0_aux.reshape(cosdir0_aux.size,1) | |
|
|
1152 | # #Minimum Distance | |
|
|
1153 | # cosDiff = (cosdir1 - cosdir0_aux)**2 | |
|
|
1154 | # indcos = cosDiff.argmin(axis = 1) | |
|
|
1155 | # #Saving Value obtained | |
|
|
1156 | # cosdir[:,i] = cosdir1[numpy.arange(len(indcos)),indcos] | |
|
|
1157 | # | |
|
|
1158 | # return cosdir0, cosdir | |
|
|
1159 | # | |
|
|
1160 | # def __calculateAOA(self, cosdir, azimuth): | |
|
|
1161 | # cosdirX = cosdir[:,0] | |
|
|
1162 | # cosdirY = cosdir[:,1] | |
|
|
1163 | # | |
|
|
1164 | # zenithAngle = numpy.arccos(numpy.sqrt(1 - cosdirX**2 - cosdirY**2))*180/numpy.pi | |
|
|
1165 | # azimuthAngle = numpy.arctan2(cosdirX,cosdirY)*180/numpy.pi + azimuth #0 deg north, 90 deg east | |
|
|
1166 | # angles = numpy.vstack((azimuthAngle, zenithAngle)).transpose() | |
|
|
1167 | # | |
|
|
1168 | # return angles | |
|
|
1169 | # | |
|
|
1170 | # def __getHeights(self, Ranges, zenith, error, minHeight, maxHeight): | |
|
|
1171 | # | |
|
|
1172 | # Ramb = 375 #Ramb = c/(2*PRF) | |
|
|
1173 | # Re = 6371 #Earth Radius | |
|
|
1174 | # heights = numpy.zeros(Ranges.shape) | |
|
|
1175 | # | |
|
|
1176 | # R_aux = numpy.array([0,1,2])*Ramb | |
|
|
1177 | # R_aux = R_aux.reshape(1,R_aux.size) | |
|
|
1178 | # | |
|
|
1179 | # Ranges = Ranges.reshape(Ranges.size,1) | |
|
|
1180 | # | |
|
|
1181 | # Ri = Ranges + R_aux | |
|
|
1182 | # hi = numpy.sqrt(Re**2 + Ri**2 + (2*Re*numpy.cos(zenith*numpy.pi/180)*Ri.transpose()).transpose()) - Re | |
|
|
1183 | # | |
|
|
1184 | # #Check if there is a height between 70 and 110 km | |
|
|
1185 | # h_bool = numpy.sum(numpy.logical_and(hi > minHeight, hi < maxHeight), axis = 1) | |
|
|
1186 | # ind_h = numpy.where(h_bool == 1)[0] | |
|
|
1187 | # | |
|
|
1188 | # hCorr = hi[ind_h, :] | |
|
|
1189 | # ind_hCorr = numpy.where(numpy.logical_and(hi > minHeight, hi < maxHeight)) | |
|
|
1190 | # | |
|
|
1191 | # hCorr = hi[ind_hCorr] | |
|
|
1192 | # heights[ind_h] = hCorr | |
|
|
1193 | # | |
|
|
1194 | # #Setting Error | |
|
|
1195 | # #Number 13: Height unresolvable echo: not valid height within 70 to 110 km | |
|
|
1196 | # #Number 14: Height ambiguous echo: more than one possible height within 70 to 110 km | |
|
|
1197 | # | |
|
|
1198 | # indInvalid2 = numpy.where(numpy.logical_and(h_bool > 1, error == 0))[0] | |
|
|
1199 | # error[indInvalid2] = 14 | |
|
|
1200 | # indInvalid1 = numpy.where(numpy.logical_and(h_bool == 0, error == 0))[0] | |
|
|
1201 | # error[indInvalid1] = 13 | |
|
|
1202 | # | |
|
|
1203 | # return heights, error | |
|
|
860 | velEst = numpy.zeros((heightList.size,2))*numpy.nan | |
|
|
861 | azimuth1 = azimuth1*numpy.pi/180 | |
|
|
862 | ||
|
|
863 | for i in range(heightList.size): | |
|
|
864 | h = heightList[i] | |
|
|
865 | indH = numpy.where((metArray1[:,2] == h)&(numpy.abs(metArray1[:,-2]) < 100))[0] | |
|
|
866 | metHeight = metArray1[indH,:] | |
|
|
867 | if metHeight.shape[0] >= 2: | |
|
|
868 | velAux = numpy.asmatrix(metHeight[:,-2]).T #Radial Velocities | |
|
|
869 | iazim = metHeight[:,1].astype(int) | |
|
|
870 | azimAux = numpy.asmatrix(azimuth1[iazim]).T #Azimuths | |
|
|
871 | A = numpy.hstack((numpy.cos(azimAux),numpy.sin(azimAux))) | |
|
|
872 | A = numpy.asmatrix(A) | |
|
|
873 | A1 = numpy.linalg.pinv(A.transpose()*A)*A.transpose() | |
|
|
874 | velHor = numpy.dot(A1,velAux) | |
|
|
875 | ||
|
|
876 | velEst[i,:] = numpy.squeeze(velHor) | |
|
|
877 | return velEst | |
|
|
1204 | 878 | |
|
|
1205 | def SpectralFitting(self, getSNR = True, path=None, file=None, groupList=None): | |
|
|
879 | def __getPhaseSlope(self, metArray, heightList, timeList): | |
|
|
880 | meteorList = [] | |
|
|
881 | #utctime sec1 height SNR velRad ph0 ph1 ph2 coh0 coh1 coh2 | |
|
|
882 | #Putting back together the meteor matrix | |
|
|
883 | utctime = metArray[:,0] | |
|
|
884 | uniqueTime = numpy.unique(utctime) | |
|
|
1206 | 885 | |
|
|
1207 | ''' | |
|
|
1208 | Function GetMoments() | |
|
|
886 | phaseDerThresh = 0.5 | |
|
|
887 | ippSeconds = timeList[1] - timeList[0] | |
|
|
888 | sec = numpy.where(timeList>1)[0][0] | |
|
|
889 | nPairs = metArray.shape[1] - 6 | |
|
|
890 | nHeights = len(heightList) | |
|
|
1209 | 891 | |
|
|
1210 | Input: | |
|
|
1211 | Output: | |
|
|
1212 | Variables modified: | |
|
|
1213 | ''' | |
|
|
1214 | if path != None: | |
|
|
1215 | sys.path.append(path) | |
|
|
1216 | self.dataOut.library = importlib.import_module(file) | |
|
|
892 | for t in uniqueTime: | |
|
|
893 | metArray1 = metArray[utctime==t,:] | |
|
|
894 | # phaseDerThresh = numpy.pi/4 #reducir Phase thresh | |
|
|
895 | tmet = metArray1[:,1].astype(int) | |
|
|
896 | hmet = metArray1[:,2].astype(int) | |
|
|
897 | ||
|
|
898 | metPhase = numpy.zeros((nPairs, heightList.size, timeList.size - 1)) | |
|
|
899 | metPhase[:,:] = numpy.nan | |
|
|
900 | metPhase[:,hmet,tmet] = metArray1[:,6:].T | |
|
|
901 | ||
|
|
902 | #Delete short trails | |
|
|
903 | metBool = ~numpy.isnan(metPhase[0,:,:]) | |
|
|
904 | heightVect = numpy.sum(metBool, axis = 1) | |
|
|
905 | metBool[heightVect<sec,:] = False | |
|
|
906 | metPhase[:,heightVect<sec,:] = numpy.nan | |
|
|
907 | ||
|
|
908 | #Derivative | |
|
|
909 | metDer = numpy.abs(metPhase[:,:,1:] - metPhase[:,:,:-1]) | |
|
|
910 | phDerAux = numpy.dstack((numpy.full((nPairs,nHeights,1), False, dtype=bool),metDer > phaseDerThresh)) | |
|
|
911 | metPhase[phDerAux] = numpy.nan | |
|
|
912 | ||
|
|
913 | #--------------------------METEOR DETECTION ----------------------------------------- | |
|
|
914 | indMet = numpy.where(numpy.any(metBool,axis=1))[0] | |
|
|
915 | ||
|
|
916 | for p in numpy.arange(nPairs): | |
|
|
917 | phase = metPhase[p,:,:] | |
|
|
918 | phDer = metDer[p,:,:] | |
|
|
919 | ||
|
|
920 | for h in indMet: | |
|
|
921 | height = heightList[h] | |
|
|
922 | phase1 = phase[h,:] #82 | |
|
|
923 | phDer1 = phDer[h,:] | |
|
|
924 | ||
|
|
925 | phase1[~numpy.isnan(phase1)] = numpy.unwrap(phase1[~numpy.isnan(phase1)]) #Unwrap | |
|
|
926 | ||
|
|
927 | indValid = numpy.where(~numpy.isnan(phase1))[0] | |
|
|
928 | initMet = indValid[0] | |
|
|
929 | endMet = 0 | |
|
|
930 | ||
|
|
931 | for i in range(len(indValid)-1): | |
|
|
932 | ||
|
|
933 | #Time difference | |
|
|
934 | inow = indValid[i] | |
|
|
935 | inext = indValid[i+1] | |
|
|
936 | idiff = inext - inow | |
|
|
937 | #Phase difference | |
|
|
938 | phDiff = numpy.abs(phase1[inext] - phase1[inow]) | |
|
|
939 | ||
|
|
940 | if idiff>sec or phDiff>numpy.pi/4 or inext==indValid[-1]: #End of Meteor | |
|
|
941 | sizeTrail = inow - initMet + 1 | |
|
|
942 | if sizeTrail>3*sec: #Too short meteors | |
|
|
943 | x = numpy.arange(initMet,inow+1)*ippSeconds | |
|
|
944 | y = phase1[initMet:inow+1] | |
|
|
945 | ynnan = ~numpy.isnan(y) | |
|
|
946 | x = x[ynnan] | |
|
|
947 | y = y[ynnan] | |
|
|
948 | slope, intercept, r_value, p_value, std_err = stats.linregress(x,y) | |
|
|
949 | ylin = x*slope + intercept | |
|
|
950 | rsq = r_value**2 | |
|
|
951 | if rsq > 0.5: | |
|
|
952 | vel = slope#*height*1000/(k*d) | |
|
|
953 | estAux = numpy.array([utctime,p,height, vel, rsq]) | |
|
|
954 | meteorList.append(estAux) | |
|
|
955 | initMet = inext | |
|
|
956 | metArray2 = numpy.array(meteorList) | |
|
|
1217 | 957 | |
|
|
1218 | #To be inserted as a parameter | |
|
|
1219 | groupArray = numpy.array(groupList) | |
|
|
1220 | # groupArray = numpy.array([[0,1],[2,3]]) | |
|
|
1221 | self.dataOut.groupList = groupArray | |
|
|
958 | return metArray2 | |
|
|
959 | ||
|
|
960 | def __calculateAzimuth1(self, rx_location, pairslist, azimuth0): | |
|
|
1222 | 961 | |
|
|
1223 | nGroups = groupArray.shape[0] | |
|
|
1224 | nChannels = self.dataIn.nChannels | |
|
|
1225 | nHeights=self.dataIn.heightList.size | |
|
|
962 | azimuth1 = numpy.zeros(len(pairslist)) | |
|
|
963 | dist = numpy.zeros(len(pairslist)) | |
|
|
1226 | 964 | |
|
|
1227 | #Parameters Array | |
|
|
1228 | self.dataOut.data_param = None | |
|
|
965 | for i in range(len(rx_location)): | |
|
|
966 | ch0 = pairslist[i][0] | |
|
|
967 | ch1 = pairslist[i][1] | |
|
|
968 | ||
|
|
969 | diffX = rx_location[ch0][0] - rx_location[ch1][0] | |
|
|
970 | diffY = rx_location[ch0][1] - rx_location[ch1][1] | |
|
|
971 | azimuth1[i] = numpy.arctan2(diffY,diffX)*180/numpy.pi | |
|
|
972 | dist[i] = numpy.sqrt(diffX**2 + diffY**2) | |
|
|
1229 | 973 | |
|
|
1230 | #Set constants | |
|
|
1231 | constants = self.dataOut.library.setConstants(self.dataIn) | |
|
|
1232 | self.dataOut.constants = constants | |
|
|
1233 | M = self.dataIn.normFactor | |
|
|
1234 | N = self.dataIn.nFFTPoints | |
|
|
1235 | ippSeconds = self.dataIn.ippSeconds | |
|
|
1236 | K = self.dataIn.nIncohInt | |
|
|
1237 | pairsArray = numpy.array(self.dataIn.pairsList) | |
|
|
974 | azimuth1 -= azimuth0 | |
|
|
975 | return azimuth1, dist | |
|
|
976 | ||
|
|
977 | def techniqueNSM_DBS(self, **kwargs): | |
|
|
978 | metArray = kwargs['metArray'] | |
|
|
979 | heightList = kwargs['heightList'] | |
|
|
980 | timeList = kwargs['timeList'] | |
|
|
981 | zenithList = kwargs['zenithList'] | |
|
|
982 | nChan = numpy.max(cmet) + 1 | |
|
|
983 | nHeights = len(heightList) | |
|
|
1238 | 984 | |
|
|
1239 | #List of possible combinations | |
|
|
1240 | listComb = itertools.combinations(numpy.arange(groupArray.shape[1]),2) | |
|
|
1241 | indCross = numpy.zeros(len(list(listComb)), dtype = 'int') | |
|
|
985 | utctime = metArray[:,0] | |
|
|
986 | cmet = metArray[:,1] | |
|
|
987 | hmet = metArray1[:,3].astype(int) | |
|
|
988 | h1met = heightList[hmet]*zenithList[cmet] | |
|
|
989 | vmet = metArray1[:,5] | |
|
|
1242 | 990 | |
|
|
1243 | if getSNR: | |
|
|
1244 | listChannels = groupArray.reshape((groupArray.size)) | |
|
|
1245 | listChannels.sort() | |
|
|
1246 | noise = self.dataIn.getNoise() | |
|
|
1247 | self.dataOut.data_SNR = self.__getSNR(self.dataIn.data_spc[listChannels,:,:], noise[listChannels]) | |
|
|
991 | for i in range(nHeights - 1): | |
|
|
992 | hmin = heightList[i] | |
|
|
993 | hmax = heightList[i + 1] | |
|
|
994 | ||
|
|
995 | vthisH = vmet[(h1met>=hmin) & (h1met<hmax)] | |
|
|
996 | ||
|
|
997 | ||
|
|
998 | ||
|
|
999 | return data_output | |
|
|
1000 | ||
|
|
1001 | def run(self, dataOut, technique, **kwargs): | |
|
|
1248 | 1002 | |
|
|
1249 | for i in range(nGroups): | |
|
|
1250 | coord = groupArray[i,:] | |
|
|
1003 | param = dataOut.data_param | |
|
|
1004 | if dataOut.abscissaList != None: | |
|
|
1005 | absc = dataOut.abscissaList[:-1] | |
|
|
1006 | noise = dataOut.noise | |
|
|
1007 | heightList = dataOut.heightList | |
|
|
1008 | SNR = dataOut.data_SNR | |
|
|
1009 | ||
|
|
1010 | if technique == 'DBS': | |
|
|
1011 | # if kwargs.has_key('dirCosx') and kwargs.has_key('dirCosy'): | |
|
|
1012 | # theta_x = numpy.array(kwargs['dirCosx']) | |
|
|
1013 | # theta_y = numpy.array(kwargs['dirCosy']) | |
|
|
1014 | # else: | |
|
|
1015 | # elev = numpy.array(kwargs['elevation']) | |
|
|
1016 | # azim = numpy.array(kwargs['azimuth']) | |
|
|
1017 | # theta_x, theta_y = self.__calculateCosDir(elev, azim) | |
|
|
1018 | # azimuth = kwargs['correctAzimuth'] | |
|
|
1019 | # if kwargs.has_key('horizontalOnly'): | |
|
|
1020 | # horizontalOnly = kwargs['horizontalOnly'] | |
|
|
1021 | # else: horizontalOnly = False | |
|
|
1022 | # if kwargs.has_key('correctFactor'): | |
|
|
1023 | # correctFactor = kwargs['correctFactor'] | |
|
|
1024 | # else: correctFactor = 1 | |
|
|
1025 | # if kwargs.has_key('channelList'): | |
|
|
1026 | # channelList = kwargs['channelList'] | |
|
|
1027 | # if len(channelList) == 2: | |
|
|
1028 | # horizontalOnly = True | |
|
|
1029 | # arrayChannel = numpy.array(channelList) | |
|
|
1030 | # param = param[arrayChannel,:,:] | |
|
|
1031 | # theta_x = theta_x[arrayChannel] | |
|
|
1032 | # theta_y = theta_y[arrayChannel] | |
|
|
1033 | ||
|
|
1034 | velRadial0 = param[:,1,:] #Radial velocity | |
|
|
1035 | # dataOut.data_output, dataOut.heightList, dataOut.data_SNR = self.techniqueDBS(velRadial0, theta_x, theta_y, azimuth, correctFactor, horizontalOnly, heightList, SNR) #DBS Function | |
|
|
1036 | dataOut.data_output, dataOut.heightList, dataOut.data_SNR = self.techniqueDBS(velRadial0, heightList, SNR, kwargs) #DBS Function | |
|
|
1037 | dataOut.utctimeInit = dataOut.utctime | |
|
|
1038 | dataOut.outputInterval = dataOut.paramInterval | |
|
|
1251 | 1039 | |
|
|
1252 | #Input data array | |
|
|
1253 | data = self.dataIn.data_spc[coord,:,:]/(M*N) | |
|
|
1254 | data = data.reshape((data.shape[0]*data.shape[1],data.shape[2])) | |
|
|
1040 | elif technique == 'SA': | |
|
|
1041 | ||
|
|
1042 | #Parameters | |
|
|
1043 | position_x = kwargs['positionX'] | |
|
|
1044 | position_y = kwargs['positionY'] | |
|
|
1045 | azimuth = kwargs['azimuth'] | |
|
|
1255 | 1046 | |
|
|
1256 | #Cross Spectra data array for Covariance Matrixes | |
|
|
1257 | ind = 0 | |
|
|
1258 | for pairs in listComb: | |
|
|
1259 | pairsSel = numpy.array([coord[x],coord[y]]) | |
|
|
1260 | indCross[ind] = int(numpy.where(numpy.all(pairsArray == pairsSel, axis = 1))[0][0]) | |
|
|
1261 | ind += 1 | |
|
|
1262 | dataCross = self.dataIn.data_cspc[indCross,:,:]/(M*N) | |
|
|
1263 | dataCross = dataCross**2/K | |
|
|
1047 | if kwargs.has_key('crosspairsList'): | |
|
|
1048 | pairs = kwargs['crosspairsList'] | |
|
|
1049 | else: | |
|
|
1050 | pairs = None | |
|
|
1051 | ||
|
|
1052 | if kwargs.has_key('correctFactor'): | |
|
|
1053 | correctFactor = kwargs['correctFactor'] | |
|
|
1054 | else: | |
|
|
1055 | correctFactor = 1 | |
|
|
1056 | ||
|
|
1057 | tau = dataOut.data_param | |
|
|
1058 | _lambda = dataOut.C/dataOut.frequency | |
|
|
1059 | pairsList = dataOut.groupList | |
|
|
1060 | nChannels = dataOut.nChannels | |
|
|
1061 | ||
|
|
1062 | dataOut.data_output = self.techniqueSA(pairs, pairsList, nChannels, tau, azimuth, _lambda, position_x, position_y, absc, correctFactor) | |
|
|
1063 | dataOut.utctimeInit = dataOut.utctime | |
|
|
1064 | dataOut.outputInterval = dataOut.timeInterval | |
|
|
1264 | 1065 | |
|
|
1265 | for h in range(nHeights): | |
|
|
1266 | # print self.dataOut.heightList[h] | |
|
|
1066 | elif technique == 'Meteors': | |
|
|
1067 | dataOut.flagNoData = True | |
|
|
1068 | self.__dataReady = False | |
|
|
1069 | ||
|
|
1070 | if kwargs.has_key('nHours'): | |
|
|
1071 | nHours = kwargs['nHours'] | |
|
|
1072 | else: | |
|
|
1073 | nHours = 1 | |
|
|
1267 | 1074 | |
|
|
1268 | #Input | |
|
|
1269 | d = data[:,h] | |
|
|
1075 | if kwargs.has_key('meteorsPerBin'): | |
|
|
1076 | meteorThresh = kwargs['meteorsPerBin'] | |
|
|
1077 | else: | |
|
|
1078 | meteorThresh = 6 | |
|
|
1079 | ||
|
|
1080 | if kwargs.has_key('hmin'): | |
|
|
1081 | hmin = kwargs['hmin'] | |
|
|
1082 | else: hmin = 70 | |
|
|
1083 | if kwargs.has_key('hmax'): | |
|
|
1084 | hmax = kwargs['hmax'] | |
|
|
1085 | else: hmax = 110 | |
|
|
1086 | ||
|
|
1087 | dataOut.outputInterval = nHours*3600 | |
|
|
1088 | ||
|
|
1089 | if self.__isConfig == False: | |
|
|
1090 | # self.__initime = dataOut.datatime.replace(minute = 0, second = 0, microsecond = 03) | |
|
|
1091 | #Get Initial LTC time | |
|
|
1092 | self.__initime = datetime.datetime.utcfromtimestamp(dataOut.utctime) | |
|
|
1093 | self.__initime = (self.__initime.replace(minute = 0, second = 0, microsecond = 0) - datetime.datetime(1970, 1, 1)).total_seconds() | |
|
|
1270 | 1094 | |
|
|
1271 | #Covariance Matrix | |
|
|
1272 | D = numpy.diag(d**2/K) | |
|
|
1273 | ind = 0 | |
|
|
1274 | for pairs in listComb: | |
|
|
1275 | #Coordinates in Covariance Matrix | |
|
|
1276 | x = pairs[0] | |
|
|
1277 | y = pairs[1] | |
|
|
1278 | #Channel Index | |
|
|
1279 | S12 = dataCross[ind,:,h] | |
|
|
1280 | D12 = numpy.diag(S12) | |
|
|
1281 | #Completing Covariance Matrix with Cross Spectras | |
|
|
1282 | D[x*N:(x+1)*N,y*N:(y+1)*N] = D12 | |
|
|
1283 | D[y*N:(y+1)*N,x*N:(x+1)*N] = D12 | |
|
|
1284 | ind += 1 | |
|
|
1285 | Dinv=numpy.linalg.inv(D) | |
|
|
1286 | L=numpy.linalg.cholesky(Dinv) | |
|
|
1287 | LT=L.T | |
|
|
1095 | self.__isConfig = True | |
|
|
1096 | ||
|
|
1097 | if self.__buffer == None: | |
|
|
1098 | self.__buffer = dataOut.data_param | |
|
|
1099 | self.__firstdata = copy.copy(dataOut) | |
|
|
1288 | 1100 | |
|
|
1289 | dp = numpy.dot(LT,d) | |
|
|
1101 | else: | |
|
|
1102 | self.__buffer = numpy.vstack((self.__buffer, dataOut.data_param)) | |
|
|
1103 | ||
|
|
1104 | self.__checkTime(dataOut.utctime, dataOut.paramInterval, dataOut.outputInterval) #Check if the buffer is ready | |
|
|
1105 | ||
|
|
1106 | if self.__dataReady: | |
|
|
1107 | dataOut.utctimeInit = self.__initime | |
|
|
1290 | 1108 | |
|
|
1291 | #Initial values | |
|
|
1292 | data_spc = self.dataIn.data_spc[coord,:,h] | |
|
|
1109 | self.__initime += dataOut.outputInterval #to erase time offset | |
|
|
1293 | 1110 | |
|
|
1294 | if (h>0)and(error1[3]<5): | |
|
|
1295 | p0 = self.dataOut.data_param[i,:,h-1] | |
|
|
1296 |
|
|
|
|
1297 | p0 = numpy.array(self.dataOut.library.initialValuesFunction(data_spc, constants, i)) | |
|
|
1111 | dataOut.data_output, dataOut.heightList = self.techniqueMeteors(self.__buffer, meteorThresh, hmin, hmax) | |
|
|
1112 | dataOut.flagNoData = False | |
|
|
1113 | self.__buffer = None | |
|
|
1298 | 1114 | |
|
|
1299 | try: | |
|
|
1300 | #Least Squares | |
|
|
1301 | minp,covp,infodict,mesg,ier = optimize.leastsq(self.__residFunction,p0,args=(dp,LT,constants),full_output=True) | |
|
|
1302 | # minp,covp = optimize.leastsq(self.__residFunction,p0,args=(dp,LT,constants)) | |
|
|
1303 | #Chi square error | |
|
|
1304 | error0 = numpy.sum(infodict['fvec']**2)/(2*N) | |
|
|
1305 | #Error with Jacobian | |
|
|
1306 | error1 = self.dataOut.library.errorFunction(minp,constants,LT) | |
|
|
1307 | except: | |
|
|
1308 | minp = p0*numpy.nan | |
|
|
1309 | error0 = numpy.nan | |
|
|
1310 | error1 = p0*numpy.nan | |
|
|
1311 | ||
|
|
1312 | #Save | |
|
|
1313 | if self.dataOut.data_param == None: | |
|
|
1314 | self.dataOut.data_param = numpy.zeros((nGroups, p0.size, nHeights))*numpy.nan | |
|
|
1315 | self.dataOut.data_error = numpy.zeros((nGroups, p0.size + 1, nHeights))*numpy.nan | |
|
|
1115 | elif technique == 'Meteors1': | |
|
|
1116 | dataOut.flagNoData = True | |
|
|
1117 | self.__dataReady = False | |
|
|
1118 | ||
|
|
1119 | if kwargs.has_key('nMins'): | |
|
|
1120 | nMins = kwargs['nMins'] | |
|
|
1121 | else: nMins = 20 | |
|
|
1122 | if kwargs.has_key('rx_location'): | |
|
|
1123 | rx_location = kwargs['rx_location'] | |
|
|
1124 | else: rx_location = [(0,1),(1,1),(1,0)] | |
|
|
1125 | if kwargs.has_key('azimuth'): | |
|
|
1126 | azimuth = kwargs['azimuth'] | |
|
|
1127 | else: azimuth = 51 | |
|
|
1128 | if kwargs.has_key('dfactor'): | |
|
|
1129 | dfactor = kwargs['dfactor'] | |
|
|
1130 | if kwargs.has_key('mode'): | |
|
|
1131 | mode = kwargs['mode'] | |
|
|
1132 | else: mode = 'SA' | |
|
|
1133 | ||
|
|
1134 | #Borrar luego esto | |
|
|
1135 | if dataOut.groupList == None: | |
|
|
1136 | dataOut.groupList = [(0,1),(0,2),(1,2)] | |
|
|
1137 | groupList = dataOut.groupList | |
|
|
1138 | C = 3e8 | |
|
|
1139 | freq = 50e6 | |
|
|
1140 | lamb = C/freq | |
|
|
1141 | k = 2*numpy.pi/lamb | |
|
|
1142 | ||
|
|
1143 | timeList = dataOut.abscissaList | |
|
|
1144 | heightList = dataOut.heightList | |
|
|
1145 | ||
|
|
1146 | if self.__isConfig == False: | |
|
|
1147 | dataOut.outputInterval = nMins*60 | |
|
|
1148 | # self.__initime = dataOut.datatime.replace(minute = 0, second = 0, microsecond = 03) | |
|
|
1149 | #Get Initial LTC time | |
|
|
1150 | initime = datetime.datetime.utcfromtimestamp(dataOut.utctime) | |
|
|
1151 | minuteAux = initime.minute | |
|
|
1152 | minuteNew = int(numpy.floor(minuteAux/nMins)*nMins) | |
|
|
1153 | self.__initime = (initime.replace(minute = minuteNew, second = 0, microsecond = 0) - datetime.datetime(1970, 1, 1)).total_seconds() | |
|
|
1154 | ||
|
|
1155 | self.__isConfig = True | |
|
|
1316 | 1156 | |
|
|
1317 | self.dataOut.data_error[i,:,h] = numpy.hstack((error0,error1)) | |
|
|
1318 |
self.dataOut.data_param |
|
|
|
1157 | if self.__buffer == None: | |
|
|
1158 | self.__buffer = dataOut.data_param | |
|
|
1159 | self.__firstdata = copy.copy(dataOut) | |
|
|
1160 | ||
|
|
1161 | else: | |
|
|
1162 | self.__buffer = numpy.vstack((self.__buffer, dataOut.data_param)) | |
|
|
1163 | ||
|
|
1164 | self.__checkTime(dataOut.utctime, dataOut.paramInterval, dataOut.outputInterval) #Check if the buffer is ready | |
|
|
1165 | ||
|
|
1166 | if self.__dataReady: | |
|
|
1167 | dataOut.utctimeInit = self.__initime | |
|
|
1168 | self.__initime += dataOut.outputInterval #to erase time offset | |
|
|
1169 | ||
|
|
1170 | metArray = self.__buffer | |
|
|
1171 | if mode == 'SA': | |
|
|
1172 | dataOut.data_output = self.techniqueNSM_SA(rx_location=rx_location, groupList=groupList, azimuth=azimuth, dfactor=dfactor, k=k,metArray=metArray, heightList=heightList,timeList=timeList) | |
|
|
1173 | elif mode == 'DBS': | |
|
|
1174 | dataOut.data_output = self.techniqueNSM_DBS(metArray=metArray,heightList=heightList,timeList=timeList) | |
|
|
1175 | dataOut.data_output = dataOut.data_output.T | |
|
|
1176 | dataOut.flagNoData = False | |
|
|
1177 | self.__buffer = None | |
|
|
1178 | ||
|
|
1179 | return | |
|
|
1180 | ||
|
|
1181 | class EWDriftsEstimation(Operation): | |
|
|
1182 | ||
|
|
1183 | def __init__(self): | |
|
|
1184 | Operation.__init__(self) | |
|
|
1185 | ||
|
|
1186 | def __correctValues(self, heiRang, phi, velRadial, SNR): | |
|
|
1187 | listPhi = phi.tolist() | |
|
|
1188 | maxid = listPhi.index(max(listPhi)) | |
|
|
1189 | minid = listPhi.index(min(listPhi)) | |
|
|
1190 | ||
|
|
1191 | rango = range(len(phi)) | |
|
|
1192 | # rango = numpy.delete(rango,maxid) | |
|
|
1193 | ||
|
|
1194 | heiRang1 = heiRang*math.cos(phi[maxid]) | |
|
|
1195 | heiRangAux = heiRang*math.cos(phi[minid]) | |
|
|
1196 | indOut = (heiRang1 < heiRangAux[0]).nonzero() | |
|
|
1197 | heiRang1 = numpy.delete(heiRang1,indOut) | |
|
|
1198 | ||
|
|
1199 | velRadial1 = numpy.zeros([len(phi),len(heiRang1)]) | |
|
|
1200 | SNR1 = numpy.zeros([len(phi),len(heiRang1)]) | |
|
|
1201 | ||
|
|
1202 | for i in rango: | |
|
|
1203 | x = heiRang*math.cos(phi[i]) | |
|
|
1204 | y1 = velRadial[i,:] | |
|
|
1205 | f1 = interpolate.interp1d(x,y1,kind = 'cubic') | |
|
|
1206 | ||
|
|
1207 | x1 = heiRang1 | |
|
|
1208 | y11 = f1(x1) | |
|
|
1209 | ||
|
|
1210 | y2 = SNR[i,:] | |
|
|
1211 | f2 = interpolate.interp1d(x,y2,kind = 'cubic') | |
|
|
1212 | y21 = f2(x1) | |
|
|
1213 | ||
|
|
1214 | velRadial1[i,:] = y11 | |
|
|
1215 | SNR1[i,:] = y21 | |
|
|
1216 | ||
|
|
1217 | return heiRang1, velRadial1, SNR1 | |
|
|
1218 | ||
|
|
1219 | def run(self, dataOut, zenith, zenithCorrection): | |
|
|
1220 | heiRang = dataOut.heightList | |
|
|
1221 | velRadial = dataOut.data_param[:,3,:] | |
|
|
1222 | SNR = dataOut.data_SNR | |
|
|
1223 | ||
|
|
1224 | zenith = numpy.array(zenith) | |
|
|
1225 | zenith -= zenithCorrection | |
|
|
1226 | zenith *= numpy.pi/180 | |
|
|
1227 | ||
|
|
1228 | heiRang1, velRadial1, SNR1 = self.__correctValues(heiRang, numpy.abs(zenith), velRadial, SNR) | |
|
|
1229 | ||
|
|
1230 | alp = zenith[0] | |
|
|
1231 | bet = zenith[1] | |
|
|
1232 | ||
|
|
1233 | w_w = velRadial1[0,:] | |
|
|
1234 | w_e = velRadial1[1,:] | |
|
|
1235 | ||
|
|
1236 | w = (w_w*numpy.sin(bet) - w_e*numpy.sin(alp))/(numpy.cos(alp)*numpy.sin(bet) - numpy.cos(bet)*numpy.sin(alp)) | |
|
|
1237 | u = (w_w*numpy.cos(bet) - w_e*numpy.cos(alp))/(numpy.sin(alp)*numpy.cos(bet) - numpy.sin(bet)*numpy.cos(alp)) | |
|
|
1238 | ||
|
|
1239 | winds = numpy.vstack((u,w)) | |
|
|
1240 | ||
|
|
1241 | dataOut.heightList = heiRang1 | |
|
|
1242 | dataOut.data_output = winds | |
|
|
1243 | dataOut.data_SNR = SNR1 | |
|
|
1244 | ||
|
|
1245 | dataOut.utctimeInit = dataOut.utctime | |
|
|
1246 | dataOut.outputInterval = dataOut.timeInterval | |
|
|
1319 | 1247 | return |
|
|
1320 | ||
|
|
1321 | def __residFunction(self, p, dp, LT, constants): | |
|
|
1322 | 1248 | |
|
|
1323 | fm = self.dataOut.library.modelFunction(p, constants) | |
|
|
1324 | fmp=numpy.dot(LT,fm) | |
|
|
1325 | ||
|
|
1326 | return dp-fmp | |
|
|
1249 | #--------------- Non Specular Meteor ---------------- | |
|
|
1327 | 1250 | |
|
|
1328 | def __getSNR(self, z, noise): | |
|
|
1329 | ||
|
|
1330 | avg = numpy.average(z, axis=1) | |
|
|
1331 | SNR = (avg.T-noise)/noise | |
|
|
1332 | SNR = SNR.T | |
|
|
1333 | return SNR | |
|
|
1334 | ||
|
|
1335 | def __chisq(p,chindex,hindex): | |
|
|
1336 | #similar to Resid but calculates CHI**2 | |
|
|
1337 | [LT,d,fm]=setupLTdfm(p,chindex,hindex) | |
|
|
1338 | dp=numpy.dot(LT,d) | |
|
|
1339 | fmp=numpy.dot(LT,fm) | |
|
|
1340 | chisq=numpy.dot((dp-fmp).T,(dp-fmp)) | |
|
|
1341 | return chisq | |
|
|
1342 | ||
|
|
1343 | def NonSpecularMeteorDetection(self, mode, SNRthresh=8, phaseDerThresh=0.5, cohThresh=0.8, allData = False): | |
|
|
1251 | class NonSpecularMeteorDetection(Operation): | |
|
|
1252 | ||
|
|
1253 | def run(self, mode, SNRthresh=8, phaseDerThresh=0.5, cohThresh=0.8, allData = False): | |
|
|
1344 | 1254 | data_acf = self.dataOut.data_pre[0] |
|
|
1345 | 1255 | data_ccf = self.dataOut.data_pre[1] |
|
|
1346 | 1256 | |
| @@ -1485,806 +1395,802 class ParametersProc(ProcessingUnit): | |||
|
|
1485 | 1395 | coordMet = numpy.where(boolMetFin) |
|
|
1486 | 1396 | |
|
|
1487 | 1397 | cmet = coordMet[0] |
|
|
1488 | tmet = coordMet[1] | |
|
|
1489 | hmet = coordMet[2] | |
|
|
1490 | ||
|
|
1491 | data_param = numpy.zeros((tmet.size, 7)) | |
|
|
1492 | data_param[:,0] = utctime | |
|
|
1493 | data_param[:,1] = cmet | |
|
|
1494 | data_param[:,2] = tmet | |
|
|
1495 | data_param[:,3] = hmet | |
|
|
1496 | data_param[:,4] = SNR[cmet,tmet,hmet].T | |
|
|
1497 | data_param[:,5] = velRad[cmet,tmet,hmet].T | |
|
|
1498 | data_param[:,6] = spcWidth[cmet,tmet,hmet].T | |
|
|
1499 | ||
|
|
1500 | # self.dataOut.data_param = data_int | |
|
|
1501 | if len(data_param) == 0: | |
|
|
1502 | self.dataOut.flagNoData = True | |
|
|
1503 | else: | |
|
|
1504 | self.dataOut.data_param = data_param | |
|
|
1505 | ||
|
|
1506 | def __erase_small(self, binArray, threshX, threshY): | |
|
|
1507 | labarray, numfeat = ndimage.measurements.label(binArray) | |
|
|
1508 | binArray1 = numpy.copy(binArray) | |
|
|
1509 | ||
|
|
1510 | for i in range(1,numfeat + 1): | |
|
|
1511 | auxBin = (labarray==i) | |
|
|
1512 | auxSize = auxBin.sum() | |
|
|
1513 | ||
|
|
1514 | x,y = numpy.where(auxBin) | |
|
|
1515 | widthX = x.max() - x.min() | |
|
|
1516 | widthY = y.max() - y.min() | |
|
|
1517 | ||
|
|
1518 | #width X: 3 seg -> 12.5*3 | |
|
|
1519 | #width Y: | |
|
|
1520 | ||
|
|
1521 | if (auxSize < 50) or (widthX < threshX) or (widthY < threshY): | |
|
|
1522 | binArray1[auxBin] = False | |
|
|
1523 | ||
|
|
1524 | return binArray1 | |
|
|
1525 | ||
|
|
1526 | def WeirdEcho(self): | |
|
|
1527 | # data_pre = self.dataOut.data_pre | |
|
|
1528 | # nHeights = self.dataOut.nHeights | |
|
|
1529 | # # nProfiles = self.dataOut.data_pre.shape[1] | |
|
|
1530 | # # data_param = numpy.zeros((len(pairslist),nProfiles,nHeights)) | |
|
|
1531 | # data_param = numpy.zeros((len(pairslist),nHeights)) | |
|
|
1532 | # | |
|
|
1533 | # for i in range(len(pairslist)): | |
|
|
1534 | # chan0 = data_pre[pairslist[i][0],:] | |
|
|
1535 | # chan1 = data_pre[pairslist[i][1],:] | |
|
|
1536 | # #calcular correlacion cruzada | |
|
|
1537 | # #magnitud es coherencia | |
|
|
1538 | # #fase es dif fase | |
|
|
1539 | # correl = chan0*numpy.conj(chan1) | |
|
|
1540 | # coherence = numpy.abs(correl)/(numpy.abs(chan0)*numpy.abs(chan1)) | |
|
|
1541 | # phase = numpy.angle(correl) | |
|
|
1542 | # # data_param[2*i,:,:] = coherence | |
|
|
1543 | # data_param[i,:] = phase | |
|
|
1544 | # | |
|
|
1545 | # self.dataOut.data_param = data_param | |
|
|
1546 | # self.dataOut.groupList = pairslist | |
|
|
1547 | data_cspc = self.dataOut.data_pre[1] | |
|
|
1548 | ccf = numpy.average(data_cspc,axis=1) | |
|
|
1549 | phases = numpy.angle(ccf).T | |
|
|
1550 | ||
|
|
1551 | meteorOps = MeteorOperations() | |
|
|
1552 | pairsList = ((0,1),(2,3)) | |
|
|
1553 | jph = numpy.array([0,0,0,0]) | |
|
|
1554 | AOAthresh = numpy.pi/8 | |
|
|
1555 | azimuth = 45 | |
|
|
1556 | error = numpy.zeros((phases.shape[0],1)) | |
|
|
1557 | AOA,error = meteorOps.getAOA(phases, pairsList, error, AOAthresh, azimuth) | |
|
|
1558 | self.dataOut.data_param = AOA.T | |
|
|
1559 | ||
|
|
1560 | class WindProfiler(Operation): | |
|
|
1561 | ||
|
|
1562 | __isConfig = False | |
|
|
1563 | ||
|
|
1564 | __initime = None | |
|
|
1565 | __lastdatatime = None | |
|
|
1566 | __integrationtime = None | |
|
|
1567 | ||
|
|
1568 | __buffer = None | |
|
|
1569 | ||
|
|
1570 | __dataReady = False | |
|
|
1571 | ||
|
|
1572 | __firstdata = None | |
|
|
1573 | ||
|
|
1574 | n = None | |
|
|
1575 | ||
|
|
1576 | def __init__(self): | |
|
|
1577 | Operation.__init__(self) | |
|
|
1578 | ||
|
|
1579 | def __calculateCosDir(self, elev, azim): | |
|
|
1580 | zen = (90 - elev)*numpy.pi/180 | |
|
|
1581 | azim = azim*numpy.pi/180 | |
|
|
1582 | cosDirX = numpy.sqrt((1-numpy.cos(zen)**2)/((1+numpy.tan(azim)**2))) | |
|
|
1583 | cosDirY = numpy.sqrt(1-numpy.cos(zen)**2-cosDirX**2) | |
|
|
1584 | ||
|
|
1585 | signX = numpy.sign(numpy.cos(azim)) | |
|
|
1586 | signY = numpy.sign(numpy.sin(azim)) | |
|
|
1587 | ||
|
|
1588 | cosDirX = numpy.copysign(cosDirX, signX) | |
|
|
1589 | cosDirY = numpy.copysign(cosDirY, signY) | |
|
|
1590 | return cosDirX, cosDirY | |
|
|
1591 | ||
|
|
1592 | def __calculateAngles(self, theta_x, theta_y, azimuth): | |
|
|
1593 | ||
|
|
1594 | dir_cosw = numpy.sqrt(1-theta_x**2-theta_y**2) | |
|
|
1595 | zenith_arr = numpy.arccos(dir_cosw) | |
|
|
1596 | azimuth_arr = numpy.arctan2(theta_x,theta_y) + azimuth*math.pi/180 | |
|
|
1597 | ||
|
|
1598 | dir_cosu = numpy.sin(azimuth_arr)*numpy.sin(zenith_arr) | |
|
|
1599 | dir_cosv = numpy.cos(azimuth_arr)*numpy.sin(zenith_arr) | |
|
|
1600 | ||
|
|
1601 | return azimuth_arr, zenith_arr, dir_cosu, dir_cosv, dir_cosw | |
|
|
1602 | ||
|
|
1603 | def __calculateMatA(self, dir_cosu, dir_cosv, dir_cosw, horOnly): | |
|
|
1604 | ||
|
|
1605 | # | |
|
|
1606 | if horOnly: | |
|
|
1607 | A = numpy.c_[dir_cosu,dir_cosv] | |
|
|
1608 | else: | |
|
|
1609 | A = numpy.c_[dir_cosu,dir_cosv,dir_cosw] | |
|
|
1610 | A = numpy.asmatrix(A) | |
|
|
1611 | A1 = numpy.linalg.inv(A.transpose()*A)*A.transpose() | |
|
|
1612 | ||
|
|
1613 | return A1 | |
|
|
1614 | ||
|
|
1615 | def __correctValues(self, heiRang, phi, velRadial, SNR): | |
|
|
1616 | listPhi = phi.tolist() | |
|
|
1617 | maxid = listPhi.index(max(listPhi)) | |
|
|
1618 | minid = listPhi.index(min(listPhi)) | |
|
|
1619 | ||
|
|
1620 | rango = range(len(phi)) | |
|
|
1621 | # rango = numpy.delete(rango,maxid) | |
|
|
1622 | ||
|
|
1623 | heiRang1 = heiRang*math.cos(phi[maxid]) | |
|
|
1624 | heiRangAux = heiRang*math.cos(phi[minid]) | |
|
|
1625 | indOut = (heiRang1 < heiRangAux[0]).nonzero() | |
|
|
1626 | heiRang1 = numpy.delete(heiRang1,indOut) | |
|
|
1627 | ||
|
|
1628 | velRadial1 = numpy.zeros([len(phi),len(heiRang1)]) | |
|
|
1629 | SNR1 = numpy.zeros([len(phi),len(heiRang1)]) | |
|
|
1630 | ||
|
|
1631 | for i in rango: | |
|
|
1632 | x = heiRang*math.cos(phi[i]) | |
|
|
1633 | y1 = velRadial[i,:] | |
|
|
1634 | f1 = interpolate.interp1d(x,y1,kind = 'cubic') | |
|
|
1635 | ||
|
|
1636 | x1 = heiRang1 | |
|
|
1637 | y11 = f1(x1) | |
|
|
1638 | ||
|
|
1639 | y2 = SNR[i,:] | |
|
|
1640 | f2 = interpolate.interp1d(x,y2,kind = 'cubic') | |
|
|
1641 | y21 = f2(x1) | |
|
|
1642 | ||
|
|
1643 | velRadial1[i,:] = y11 | |
|
|
1644 | SNR1[i,:] = y21 | |
|
|
1645 | ||
|
|
1646 | return heiRang1, velRadial1, SNR1 | |
|
|
1647 | ||
|
|
1648 | def __calculateVelUVW(self, A, velRadial): | |
|
|
1649 | ||
|
|
1650 | #Operacion Matricial | |
|
|
1651 | # velUVW = numpy.zeros((velRadial.shape[1],3)) | |
|
|
1652 | # for ind in range(velRadial.shape[1]): | |
|
|
1653 | # velUVW[ind,:] = numpy.dot(A,velRadial[:,ind]) | |
|
|
1654 | # velUVW = velUVW.transpose() | |
|
|
1655 | velUVW = numpy.zeros((A.shape[0],velRadial.shape[1])) | |
|
|
1656 | velUVW[:,:] = numpy.dot(A,velRadial) | |
|
|
1657 | ||
|
|
1658 | ||
|
|
1659 | return velUVW | |
|
|
1660 | ||
|
|
1661 | def techniqueDBS(self, velRadial0, dirCosx, disrCosy, azimuth, correct, horizontalOnly, heiRang, SNR0): | |
|
|
1662 | """ | |
|
|
1663 | Function that implements Doppler Beam Swinging (DBS) technique. | |
|
|
1664 | ||
|
|
1665 | Input: Radial velocities, Direction cosines (x and y) of the Beam, Antenna azimuth, | |
|
|
1666 | Direction correction (if necessary), Ranges and SNR | |
|
|
1667 | ||
|
|
1668 | Output: Winds estimation (Zonal, Meridional and Vertical) | |
|
|
1669 | ||
|
|
1670 | Parameters affected: Winds, height range, SNR | |
|
|
1671 | """ | |
|
|
1672 | azimuth_arr, zenith_arr, dir_cosu, dir_cosv, dir_cosw = self.__calculateAngles(dirCosx, disrCosy, azimuth) | |
|
|
1673 | heiRang1, velRadial1, SNR1 = self.__correctValues(heiRang, zenith_arr, correct*velRadial0, SNR0) | |
|
|
1674 | A = self.__calculateMatA(dir_cosu, dir_cosv, dir_cosw, horizontalOnly) | |
|
|
1675 | ||
|
|
1676 | #Calculo de Componentes de la velocidad con DBS | |
|
|
1677 | winds = self.__calculateVelUVW(A,velRadial1) | |
|
|
1678 | ||
|
|
1679 | return winds, heiRang1, SNR1 | |
|
|
1680 | ||
|
|
1681 | def __calculateDistance(self, posx, posy, pairsCrossCorr, pairsList, pairs, azimuth = None): | |
|
|
1682 | ||
|
|
1683 | posx = numpy.asarray(posx) | |
|
|
1684 | posy = numpy.asarray(posy) | |
|
|
1685 | ||
|
|
1686 | #Rotacion Inversa para alinear con el azimuth | |
|
|
1687 | if azimuth!= None: | |
|
|
1688 | azimuth = azimuth*math.pi/180 | |
|
|
1689 | posx1 = posx*math.cos(azimuth) + posy*math.sin(azimuth) | |
|
|
1690 | posy1 = -posx*math.sin(azimuth) + posy*math.cos(azimuth) | |
|
|
1691 | else: | |
|
|
1692 | posx1 = posx | |
|
|
1693 | posy1 = posy | |
|
|
1694 | ||
|
|
1695 | #Calculo de Distancias | |
|
|
1696 | distx = numpy.zeros(pairsCrossCorr.size) | |
|
|
1697 | disty = numpy.zeros(pairsCrossCorr.size) | |
|
|
1698 | dist = numpy.zeros(pairsCrossCorr.size) | |
|
|
1699 | ang = numpy.zeros(pairsCrossCorr.size) | |
|
|
1700 | ||
|
|
1701 | for i in range(pairsCrossCorr.size): | |
|
|
1702 | distx[i] = posx1[pairsList[pairsCrossCorr[i]][1]] - posx1[pairsList[pairsCrossCorr[i]][0]] | |
|
|
1703 | disty[i] = posy1[pairsList[pairsCrossCorr[i]][1]] - posy1[pairsList[pairsCrossCorr[i]][0]] | |
|
|
1704 | dist[i] = numpy.sqrt(distx[i]**2 + disty[i]**2) | |
|
|
1705 | ang[i] = numpy.arctan2(disty[i],distx[i]) | |
|
|
1706 | #Calculo de Matrices | |
|
|
1707 | nPairs = len(pairs) | |
|
|
1708 | ang1 = numpy.zeros((nPairs, 2, 1)) | |
|
|
1709 | dist1 = numpy.zeros((nPairs, 2, 1)) | |
|
|
1710 | ||
|
|
1711 | for j in range(nPairs): | |
|
|
1712 | dist1[j,0,0] = dist[pairs[j][0]] | |
|
|
1713 | dist1[j,1,0] = dist[pairs[j][1]] | |
|
|
1714 | ang1[j,0,0] = ang[pairs[j][0]] | |
|
|
1715 | ang1[j,1,0] = ang[pairs[j][1]] | |
|
|
1398 | tmet = coordMet[1] | |
|
|
1399 | hmet = coordMet[2] | |
|
|
1716 | 1400 | |
|
|
1717 | return distx,disty, dist1,ang1 | |
|
|
1718 | ||
|
|
1719 | def __calculateVelVer(self, phase, lagTRange, _lambda): | |
|
|
1401 | data_param = numpy.zeros((tmet.size, 7)) | |
|
|
1402 | data_param[:,0] = utctime | |
|
|
1403 | data_param[:,1] = cmet | |
|
|
1404 | data_param[:,2] = tmet | |
|
|
1405 | data_param[:,3] = hmet | |
|
|
1406 | data_param[:,4] = SNR[cmet,tmet,hmet].T | |
|
|
1407 | data_param[:,5] = velRad[cmet,tmet,hmet].T | |
|
|
1408 | data_param[:,6] = spcWidth[cmet,tmet,hmet].T | |
|
|
1409 | ||
|
|
1410 | # self.dataOut.data_param = data_int | |
|
|
1411 | if len(data_param) == 0: | |
|
|
1412 | self.dataOut.flagNoData = True | |
|
|
1413 | else: | |
|
|
1414 | self.dataOut.data_param = data_param | |
|
|
1720 | 1415 | |
|
|
1721 | Ts = lagTRange[1] - lagTRange[0] | |
|
|
1722 | velW = -_lambda*phase/(4*math.pi*Ts) | |
|
|
1723 | ||
|
|
1724 | return velW | |
|
|
1725 | ||
|
|
1726 | def __calculateVelHorDir(self, dist, tau1, tau2, ang): | |
|
|
1727 | nPairs = tau1.shape[0] | |
|
|
1728 | vel = numpy.zeros((nPairs,3,tau1.shape[2])) | |
|
|
1416 | def __erase_small(self, binArray, threshX, threshY): | |
|
|
1417 | labarray, numfeat = ndimage.measurements.label(binArray) | |
|
|
1418 | binArray1 = numpy.copy(binArray) | |
|
|
1729 | 1419 | |
|
|
1730 | angCos = numpy.cos(ang) | |
|
|
1731 | angSin = numpy.sin(ang) | |
|
|
1420 | for i in range(1,numfeat + 1): | |
|
|
1421 | auxBin = (labarray==i) | |
|
|
1422 | auxSize = auxBin.sum() | |
|
|
1423 | ||
|
|
1424 | x,y = numpy.where(auxBin) | |
|
|
1425 | widthX = x.max() - x.min() | |
|
|
1426 | widthY = y.max() - y.min() | |
|
|
1427 | ||
|
|
1428 | #width X: 3 seg -> 12.5*3 | |
|
|
1429 | #width Y: | |
|
|
1430 | ||
|
|
1431 | if (auxSize < 50) or (widthX < threshX) or (widthY < threshY): | |
|
|
1432 | binArray1[auxBin] = False | |
|
|
1433 | ||
|
|
1434 | return binArray1 | |
|
|
1435 | ||
|
|
1436 | #--------------- Specular Meteor ---------------- | |
|
|
1437 | ||
|
|
1438 | class MeteorDetection(Operation): | |
|
|
1439 | ''' | |
|
|
1440 | Function DetectMeteors() | |
|
|
1441 | Project developed with paper: | |
|
|
1442 | HOLDSWORTH ET AL. 2004 | |
|
|
1443 | ||
|
|
1444 | Input: | |
|
|
1445 | self.dataOut.data_pre | |
|
|
1732 | 1446 |
|
|
|
1733 | vel0 = dist*tau1/(2*tau2**2) | |
|
|
1734 | vel[:,0,:] = (vel0*angCos).sum(axis = 1) | |
|
|
1735 | vel[:,1,:] = (vel0*angSin).sum(axis = 1) | |
|
|
1447 | centerReceiverIndex: From the channels, which is the center receiver | |
|
|
1448 | ||
|
|
1449 | hei_ref: Height reference for the Beacon signal extraction | |
|
|
1450 | tauindex: | |
|
|
1451 | predefinedPhaseShifts: Predefined phase offset for the voltge signals | |
|
|
1452 | ||
|
|
1453 | cohDetection: Whether to user Coherent detection or not | |
|
|
1454 | cohDet_timeStep: Coherent Detection calculation time step | |
|
|
1455 | cohDet_thresh: Coherent Detection phase threshold to correct phases | |
|
|
1456 | ||
|
|
1457 | noise_timeStep: Noise calculation time step | |
|
|
1458 | noise_multiple: Noise multiple to define signal threshold | |
|
|
1459 | ||
|
|
1460 | multDet_timeLimit: Multiple Detection Removal time limit in seconds | |
|
|
1461 | multDet_rangeLimit: Multiple Detection Removal range limit in km | |
|
|
1462 | ||
|
|
1463 | phaseThresh: Maximum phase difference between receiver to be consider a meteor | |
|
|
1464 | SNRThresh: Minimum SNR threshold of the meteor signal to be consider a meteor | |
|
|
1465 | ||
|
|
1466 | hmin: Minimum Height of the meteor to use it in the further wind estimations | |
|
|
1467 | hmax: Maximum Height of the meteor to use it in the further wind estimations | |
|
|
1468 | azimuth: Azimuth angle correction | |
|
|
1469 | ||
|
|
1470 | Affected: | |
|
|
1471 | self.dataOut.data_param | |
|
|
1736 | 1472 |
|
|
|
1737 | ind = numpy.where(numpy.isinf(vel)) | |
|
|
1738 | vel[ind] = numpy.nan | |
|
|
1739 | ||
|
|
1740 | return vel | |
|
|
1741 | ||
|
|
1742 | def __getPairsAutoCorr(self, pairsList, nChannels): | |
|
|
1743 | ||
|
|
1744 | pairsAutoCorr = numpy.zeros(nChannels, dtype = 'int')*numpy.nan | |
|
|
1473 | Rejection Criteria (Errors): | |
|
|
1474 | 0: No error; analysis OK | |
|
|
1475 | 1: SNR < SNR threshold | |
|
|
1476 | 2: angle of arrival (AOA) ambiguously determined | |
|
|
1477 | 3: AOA estimate not feasible | |
|
|
1478 | 4: Large difference in AOAs obtained from different antenna baselines | |
|
|
1479 | 5: echo at start or end of time series | |
|
|
1480 | 6: echo less than 5 examples long; too short for analysis | |
|
|
1481 | 7: echo rise exceeds 0.3s | |
|
|
1482 | 8: echo decay time less than twice rise time | |
|
|
1483 | 9: large power level before echo | |
|
|
1484 | 10: large power level after echo | |
|
|
1485 | 11: poor fit to amplitude for estimation of decay time | |
|
|
1486 | 12: poor fit to CCF phase variation for estimation of radial drift velocity | |
|
|
1487 | 13: height unresolvable echo: not valid height within 70 to 110 km | |
|
|
1488 | 14: height ambiguous echo: more then one possible height within 70 to 110 km | |
|
|
1489 | 15: radial drift velocity or projected horizontal velocity exceeds 200 m/s | |
|
|
1490 | 16: oscilatory echo, indicating event most likely not an underdense echo | |
|
|
1745 | 1491 |
|
|
|
1746 | for l in range(len(pairsList)): | |
|
|
1747 | firstChannel = pairsList[l][0] | |
|
|
1748 | secondChannel = pairsList[l][1] | |
|
|
1749 | ||
|
|
1750 | #Obteniendo pares de Autocorrelacion | |
|
|
1751 | if firstChannel == secondChannel: | |
|
|
1752 | pairsAutoCorr[firstChannel] = int(l) | |
|
|
1753 | ||
|
|
1754 | pairsAutoCorr = pairsAutoCorr.astype(int) | |
|
|
1492 | 17: phase difference in meteor Reestimation | |
|
|
1755 | 1493 |
|
|
|
1756 | pairsCrossCorr = range(len(pairsList)) | |
|
|
1757 | pairsCrossCorr = numpy.delete(pairsCrossCorr,pairsAutoCorr) | |
|
|
1494 | Data Storage: | |
|
|
1495 | Meteors for Wind Estimation (8): | |
|
|
1496 | Utc Time | Range Height | |
|
|
1497 | Azimuth Zenith errorCosDir | |
|
|
1498 | VelRad errorVelRad | |
|
|
1499 | Phase0 Phase1 Phase2 Phase3 | |
|
|
1500 | TypeError | |
|
|
1758 | 1501 |
|
|
|
1759 | return pairsAutoCorr, pairsCrossCorr | |
|
|
1502 | ''' | |
|
|
1760 | 1503 | |
|
|
1761 | def techniqueSA(self, pairsSelected, pairsList, nChannels, tau, azimuth, _lambda, position_x, position_y, lagTRange, correctFactor): | |
|
|
1762 | """ | |
|
|
1763 | Function that implements Spaced Antenna (SA) technique. | |
|
|
1764 | ||
|
|
1765 | Input: Radial velocities, Direction cosines (x and y) of the Beam, Antenna azimuth, | |
|
|
1766 | Direction correction (if necessary), Ranges and SNR | |
|
|
1504 | def run(self, dataOut, hei_ref = None, tauindex = 0, | |
|
|
1505 | phaseOffsets = None, | |
|
|
1506 | cohDetection = False, cohDet_timeStep = 1, cohDet_thresh = 25, | |
|
|
1507 | noise_timeStep = 4, noise_multiple = 4, | |
|
|
1508 | multDet_timeLimit = 1, multDet_rangeLimit = 3, | |
|
|
1509 | phaseThresh = 20, SNRThresh = 5, | |
|
|
1510 | hmin = 50, hmax=150, azimuth = 0, | |
|
|
1511 | channelPositions = None) : | |
|
|
1767 | 1512 | |
|
|
1768 | Output: Winds estimation (Zonal, Meridional and Vertical) | |
|
|
1513 | ||
|
|
1514 | #Getting Pairslist | |
|
|
1515 | if channelPositions == None: | |
|
|
1516 | # channelPositions = [(2.5,0), (0,2.5), (0,0), (0,4.5), (-2,0)] #T | |
|
|
1517 | channelPositions = [(4.5,2), (2,4.5), (2,2), (2,0), (0,2)] #Estrella | |
|
|
1518 | meteorOps = MeteorOperations() | |
|
|
1519 | pairslist0, distances = meteorOps.getPhasePairs(channelPositions) | |
|
|
1769 | 1520 | |
|
|
1770 | Parameters affected: Winds | |
|
|
1771 | """ | |
|
|
1772 | #Cross Correlation pairs obtained | |
|
|
1773 | pairsAutoCorr, pairsCrossCorr = self.__getPairsAutoCorr(pairsList, nChannels) | |
|
|
1774 | pairsArray = numpy.array(pairsList)[pairsCrossCorr] | |
|
|
1775 | pairsSelArray = numpy.array(pairsSelected) | |
|
|
1776 | pairs = [] | |
|
|
1521 | #Get Beacon signal | |
|
|
1522 | newheis = numpy.where(self.dataOut.heightList>self.dataOut.radarControllerHeaderObj.Taus[tauindex]) | |
|
|
1777 | 1523 | |
|
|
1778 | #Wind estimation pairs obtained | |
|
|
1779 | for i in range(pairsSelArray.shape[0]/2): | |
|
|
1780 | ind1 = numpy.where(numpy.all(pairsArray == pairsSelArray[2*i], axis = 1))[0][0] | |
|
|
1781 | ind2 = numpy.where(numpy.all(pairsArray == pairsSelArray[2*i + 1], axis = 1))[0][0] | |
|
|
1782 | pairs.append((ind1,ind2)) | |
|
|
1524 | if hei_ref != None: | |
|
|
1525 | newheis = numpy.where(self.dataOut.heightList>hei_ref) | |
|
|
1783 | 1526 | |
|
|
1784 | indtau = tau.shape[0]/2 | |
|
|
1785 | tau1 = tau[:indtau,:] | |
|
|
1786 | tau2 = tau[indtau:-1,:] | |
|
|
1787 | tau1 = tau1[pairs,:] | |
|
|
1788 | tau2 = tau2[pairs,:] | |
|
|
1789 | phase1 = tau[-1,:] | |
|
|
1527 | heiRang = dataOut.getHeiRange() | |
|
|
1528 | ||
|
|
1529 | #****************REMOVING HARDWARE PHASE DIFFERENCES*************** | |
|
|
1530 | # see if the user put in pre defined phase shifts | |
|
|
1531 | voltsPShift = self.dataOut.data_pre.copy() | |
|
|
1790 | 1532 | |
|
|
1791 | #--------------------------------------------------------------------- | |
|
|
1792 | #Metodo Directo | |
|
|
1793 | distx, disty, dist, ang = self.__calculateDistance(position_x, position_y, pairsCrossCorr, pairsList, pairs,azimuth) | |
|
|
1794 | winds = self.__calculateVelHorDir(dist, tau1, tau2, ang) | |
|
|
1795 | winds = stats.nanmean(winds, axis=0) | |
|
|
1796 | #--------------------------------------------------------------------- | |
|
|
1797 | #Metodo General | |
|
|
1798 | # distx, disty, dist = self.calculateDistance(position_x,position_y,pairsCrossCorr, pairsList, azimuth) | |
|
|
1799 | # #Calculo Coeficientes de Funcion de Correlacion | |
|
|
1800 | # F,G,A,B,H = self.calculateCoef(tau1,tau2,distx,disty,n) | |
|
|
1801 | # #Calculo de Velocidades | |
|
|
1802 | # winds = self.calculateVelUV(F,G,A,B,H) | |
|
|
1533 | # if predefinedPhaseShifts != None: | |
|
|
1534 | # hardwarePhaseShifts = numpy.array(predefinedPhaseShifts)*numpy.pi/180 | |
|
|
1535 | # | |
|
|
1536 | # # elif beaconPhaseShifts: | |
|
|
1537 | # # #get hardware phase shifts using beacon signal | |
|
|
1538 | # # hardwarePhaseShifts = self.__getHardwarePhaseDiff(self.dataOut.data_pre, pairslist, newheis, 10) | |
|
|
1539 | # # hardwarePhaseShifts = numpy.insert(hardwarePhaseShifts,centerReceiverIndex,0) | |
|
|
1540 | # | |
|
|
1541 | # else: | |
|
|
1542 | # hardwarePhaseShifts = numpy.zeros(5) | |
|
|
1543 | # | |
|
|
1544 | # voltsPShift = numpy.zeros((self.dataOut.data_pre.shape[0],self.dataOut.data_pre.shape[1],self.dataOut.data_pre.shape[2]), dtype = 'complex') | |
|
|
1545 | # for i in range(self.dataOut.data_pre.shape[0]): | |
|
|
1546 | # voltsPShift[i,:,:] = self.__shiftPhase(self.dataOut.data_pre[i,:,:], hardwarePhaseShifts[i]) | |
|
|
1803 | 1547 | |
|
|
1804 | #--------------------------------------------------------------------- | |
|
|
1805 | winds[2,:] = self.__calculateVelVer(phase1, lagTRange, _lambda) | |
|
|
1806 | winds = correctFactor*winds | |
|
|
1807 | return winds | |
|
|
1808 | ||
|
|
1809 | def __checkTime(self, currentTime, paramInterval, outputInterval): | |
|
|
1810 | ||
|
|
1811 | dataTime = currentTime + paramInterval | |
|
|
1812 | deltaTime = dataTime - self.__initime | |
|
|
1813 | ||
|
|
1814 | if deltaTime >= outputInterval or deltaTime < 0: | |
|
|
1815 | self.__dataReady = True | |
|
|
1816 | return | |
|
|
1548 | #******************END OF REMOVING HARDWARE PHASE DIFFERENCES********* | |
|
|
1817 | 1549 | |
|
|
1818 | def techniqueMeteors(self, arrayMeteor, meteorThresh, heightMin, heightMax): | |
|
|
1819 | ''' | |
|
|
1820 | Function that implements winds estimation technique with detected meteors. | |
|
|
1821 | ||
|
|
1822 | Input: Detected meteors, Minimum meteor quantity to wind estimation | |
|
|
1823 | ||
|
|
1824 | Output: Winds estimation (Zonal and Meridional) | |
|
|
1825 | ||
|
|
1826 | Parameters affected: Winds | |
|
|
1827 | ''' | |
|
|
1828 | # print arrayMeteor.shape | |
|
|
1829 | #Settings | |
|
|
1830 | nInt = (heightMax - heightMin)/2 | |
|
|
1831 | # print nInt | |
|
|
1832 | nInt = int(nInt) | |
|
|
1833 | # print nInt | |
|
|
1834 | winds = numpy.zeros((2,nInt))*numpy.nan | |
|
|
1835 | ||
|
|
1836 | #Filter errors | |
|
|
1837 | error = numpy.where(arrayMeteor[:,-1] == 0)[0] | |
|
|
1838 | finalMeteor = arrayMeteor[error,:] | |
|
|
1550 | #Remove DC | |
|
|
1551 | voltsDC = numpy.mean(voltsPShift,1) | |
|
|
1552 | voltsDC = numpy.mean(voltsDC,1) | |
|
|
1553 | for i in range(voltsDC.shape[0]): | |
|
|
1554 | voltsPShift[i] = voltsPShift[i] - voltsDC[i] | |
|
|
1555 | ||
|
|
1556 | #Don't considerate last heights, theyre used to calculate Hardware Phase Shift | |
|
|
1557 | voltsPShift = voltsPShift[:,:,:newheis[0][0]] | |
|
|
1839 | 1558 | |
|
|
1840 | #Meteor Histogram | |
|
|
1841 | finalHeights = finalMeteor[:,2] | |
|
|
1842 | hist = numpy.histogram(finalHeights, bins = nInt, range = (heightMin,heightMax)) | |
|
|
1843 | nMeteorsPerI = hist[0] | |
|
|
1844 | heightPerI = hist[1] | |
|
|
1559 | #************ FIND POWER OF DATA W/COH OR NON COH DETECTION (3.4) ********** | |
|
|
1560 | #Coherent Detection | |
|
|
1561 | if cohDetection: | |
|
|
1562 | #use coherent detection to get the net power | |
|
|
1563 | cohDet_thresh = cohDet_thresh*numpy.pi/180 | |
|
|
1564 | voltsPShift = self.__coherentDetection(voltsPShift, cohDet_timeStep, self.dataOut.timeInterval, pairslist0, cohDet_thresh) | |
|
|
1845 | 1565 | |
|
|
1846 |
# |
|
|
|
1847 | indSort = finalHeights.argsort() | |
|
|
1848 | finalMeteor2 = finalMeteor[indSort,:] | |
|
|
1566 | #Non-coherent detection! | |
|
|
1567 | powerNet = numpy.nansum(numpy.abs(voltsPShift[:,:,:])**2,0) | |
|
|
1568 | #********** END OF COH/NON-COH POWER CALCULATION********************** | |
|
|
1569 | ||
|
|
1570 | #********** FIND THE NOISE LEVEL AND POSSIBLE METEORS **************** | |
|
|
1571 | #Get noise | |
|
|
1572 | noise, noise1 = self.__getNoise(powerNet, noise_timeStep, self.dataOut.timeInterval) | |
|
|
1573 | # noise = self.getNoise1(powerNet, noise_timeStep, self.dataOut.timeInterval) | |
|
|
1574 | #Get signal threshold | |
|
|
1575 | signalThresh = noise_multiple*noise | |
|
|
1576 | #Meteor echoes detection | |
|
|
1577 | listMeteors = self.__findMeteors(powerNet, signalThresh) | |
|
|
1578 | #******* END OF NOISE LEVEL AND POSSIBLE METEORS CACULATION ********** | |
|
|
1849 | 1579 | |
|
|
1850 | # Calculating winds | |
|
|
1851 | ind1 = 0 | |
|
|
1852 | ind2 = 0 | |
|
|
1580 | #************** REMOVE MULTIPLE DETECTIONS (3.5) *************************** | |
|
|
1581 | #Parameters | |
|
|
1582 | heiRange = self.dataOut.getHeiRange() | |
|
|
1583 | rangeInterval = heiRange[1] - heiRange[0] | |
|
|
1584 | rangeLimit = multDet_rangeLimit/rangeInterval | |
|
|
1585 | timeLimit = multDet_timeLimit/self.dataOut.timeInterval | |
|
|
1586 | #Multiple detection removals | |
|
|
1587 | listMeteors1 = self.__removeMultipleDetections(listMeteors, rangeLimit, timeLimit) | |
|
|
1588 | #************ END OF REMOVE MULTIPLE DETECTIONS ********************** | |
|
|
1853 | 1589 | |
|
|
1854 | for i in range(nInt): | |
|
|
1855 | nMet = nMeteorsPerI[i] | |
|
|
1856 | ind1 = ind2 | |
|
|
1857 | ind2 = ind1 + nMet | |
|
|
1590 | #********************* METEOR REESTIMATION (3.7, 3.8, 3.9, 3.10) ******************** | |
|
|
1591 | #Parameters | |
|
|
1592 | phaseThresh = phaseThresh*numpy.pi/180 | |
|
|
1593 | thresh = [phaseThresh, noise_multiple, SNRThresh] | |
|
|
1594 | #Meteor reestimation (Errors N 1, 6, 12, 17) | |
|
|
1595 | listMeteors2, listMeteorsPower, listMeteorsVolts = self.__meteorReestimation(listMeteors1, voltsPShift, pairslist0, thresh, noise, self.dataOut.timeInterval, self.dataOut.frequency) | |
|
|
1596 | # listMeteors2, listMeteorsPower, listMeteorsVolts = self.meteorReestimation3(listMeteors2, listMeteorsPower, listMeteorsVolts, voltsPShift, pairslist, thresh, noise) | |
|
|
1597 | #Estimation of decay times (Errors N 7, 8, 11) | |
|
|
1598 | listMeteors3 = self.__estimateDecayTime(listMeteors2, listMeteorsPower, self.dataOut.timeInterval, self.dataOut.frequency) | |
|
|
1599 | #******************* END OF METEOR REESTIMATION ******************* | |
|
|
1600 | ||
|
|
1601 | #********************* METEOR PARAMETERS CALCULATION (3.11, 3.12, 3.13) ************************** | |
|
|
1602 | #Calculating Radial Velocity (Error N 15) | |
|
|
1603 | radialStdThresh = 10 | |
|
|
1604 | listMeteors4 = self.__getRadialVelocity(listMeteors3, listMeteorsVolts, radialStdThresh, pairslist0, self.dataOut.timeInterval) | |
|
|
1605 | ||
|
|
1606 | if len(listMeteors4) > 0: | |
|
|
1607 | #Setting New Array | |
|
|
1608 | date = self.dataOut.utctime | |
|
|
1609 | arrayParameters = self.__setNewArrays(listMeteors4, date, heiRang) | |
|
|
1858 | 1610 | |
|
|
1859 | meteorAux = finalMeteor2[ind1:ind2,:] | |
|
|
1611 | #Correcting phase offset | |
|
|
1612 | if phaseOffsets != None: | |
|
|
1613 | phaseOffsets = numpy.array(phaseOffsets)*numpy.pi/180 | |
|
|
1614 | arrayParameters[:,8:12] = numpy.unwrap(arrayParameters[:,8:12] + phaseOffsets) | |
|
|
1860 | 1615 | |
|
|
1861 | if meteorAux.shape[0] >= meteorThresh: | |
|
|
1862 | vel = meteorAux[:, 6] | |
|
|
1863 | zen = meteorAux[:, 4]*numpy.pi/180 | |
|
|
1864 | azim = meteorAux[:, 3]*numpy.pi/180 | |
|
|
1865 | ||
|
|
1866 | n = numpy.cos(zen) | |
|
|
1867 | # m = (1 - n**2)/(1 - numpy.tan(azim)**2) | |
|
|
1868 | # l = m*numpy.tan(azim) | |
|
|
1869 | l = numpy.sin(zen)*numpy.sin(azim) | |
|
|
1870 | m = numpy.sin(zen)*numpy.cos(azim) | |
|
|
1871 |
|
|
|
|
1872 | A = numpy.vstack((l, m)).transpose() | |
|
|
1873 | A1 = numpy.dot(numpy.linalg.inv( numpy.dot(A.transpose(),A) ),A.transpose()) | |
|
|
1874 | windsAux = numpy.dot(A1, vel) | |
|
|
1875 | ||
|
|
1876 | winds[0,i] = windsAux[0] | |
|
|
1877 | winds[1,i] = windsAux[1] | |
|
|
1616 | #Second Pairslist | |
|
|
1617 | pairsList = [] | |
|
|
1618 | pairx = (0,1) | |
|
|
1619 | pairy = (2,3) | |
|
|
1620 | pairsList.append(pairx) | |
|
|
1621 | pairsList.append(pairy) | |
|
|
1622 | ||
|
|
1623 | jph = numpy.array([0,0,0,0]) | |
|
|
1624 | h = (hmin,hmax) | |
|
|
1625 | arrayParameters = meteorOps.getMeteorParams(arrayParameters, azimuth, h, pairsList, distances, jph) | |
|
|
1626 | ||
|
|
1627 | # #Calculate AOA (Error N 3, 4) | |
|
|
1628 | # #JONES ET AL. 1998 | |
|
|
1629 | # error = arrayParameters[:,-1] | |
|
|
1630 | # AOAthresh = numpy.pi/8 | |
|
|
1631 | # phases = -arrayParameters[:,9:13] | |
|
|
1632 | # arrayParameters[:,4:7], arrayParameters[:,-1] = meteorOps.getAOA(phases, pairsList, error, AOAthresh, azimuth) | |
|
|
1633 | # | |
|
|
1634 | # #Calculate Heights (Error N 13 and 14) | |
|
|
1635 | # error = arrayParameters[:,-1] | |
|
|
1636 | # Ranges = arrayParameters[:,2] | |
|
|
1637 | # zenith = arrayParameters[:,5] | |
|
|
1638 | # arrayParameters[:,3], arrayParameters[:,-1] = meteorOps.getHeights(Ranges, zenith, error, hmin, hmax) | |
|
|
1639 | # error = arrayParameters[:,-1] | |
|
|
1640 | #********************* END OF PARAMETERS CALCULATION ************************** | |
|
|
1878 | 1641 | |
|
|
1879 | return winds, heightPerI[:-1] | |
|
|
1880 | ||
|
|
1881 | def techniqueNSM_SA(self, **kwargs): | |
|
|
1882 | metArray = kwargs['metArray'] | |
|
|
1883 | heightList = kwargs['heightList'] | |
|
|
1884 | timeList = kwargs['timeList'] | |
|
|
1642 | #***************************+ PASS DATA TO NEXT STEP ********************** | |
|
|
1643 | # arrayFinal = arrayParameters.reshape((1,arrayParameters.shape[0],arrayParameters.shape[1])) | |
|
|
1644 | self.dataOut.data_param = arrayParameters | |
|
|
1645 | ||
|
|
1646 | if arrayParameters == None: | |
|
|
1647 | self.dataOut.flagNoData = True | |
|
|
1648 | else: | |
|
|
1649 | self.dataOut.flagNoData = True | |
|
|
1650 | ||
|
|
1651 | return | |
|
|
1885 | 1652 | |
|
|
1886 | rx_location = kwargs['rx_location'] | |
|
|
1887 | groupList = kwargs['groupList'] | |
|
|
1888 | azimuth = kwargs['azimuth'] | |
|
|
1889 | dfactor = kwargs['dfactor'] | |
|
|
1890 | k = kwargs['k'] | |
|
|
1653 | def __getHardwarePhaseDiff(self, voltage0, pairslist, newheis, n): | |
|
|
1654 | ||
|
|
1655 | minIndex = min(newheis[0]) | |
|
|
1656 | maxIndex = max(newheis[0]) | |
|
|
1891 | 1657 | |
|
|
1892 | azimuth1, dist = self.__calculateAzimuth1(rx_location, groupList, azimuth) | |
|
|
1893 | d = dist*dfactor | |
|
|
1894 | #Phase calculation | |
|
|
1895 | metArray1 = self.__getPhaseSlope(metArray, heightList, timeList) | |
|
|
1658 | voltage = voltage0[:,:,minIndex:maxIndex+1] | |
|
|
1659 | nLength = voltage.shape[1]/n | |
|
|
1660 | nMin = 0 | |
|
|
1661 | nMax = 0 | |
|
|
1662 | phaseOffset = numpy.zeros((len(pairslist),n)) | |
|
|
1896 | 1663 | |
|
|
1897 | metArray1[:,-2] = metArray1[:,-2]*metArray1[:,2]*1000/(k*d[metArray1[:,1].astype(int)]) #angles into velocities | |
|
|
1664 | for i in range(n): | |
|
|
1665 | nMax += nLength | |
|
|
1666 | phaseCCF = -numpy.angle(self.__calculateCCF(voltage[:,nMin:nMax,:], pairslist, [0])) | |
|
|
1667 | phaseCCF = numpy.mean(phaseCCF, axis = 2) | |
|
|
1668 | phaseOffset[:,i] = phaseCCF.transpose() | |
|
|
1669 | nMin = nMax | |
|
|
1670 | # phaseDiff, phaseArrival = self.estimatePhaseDifference(voltage, pairslist) | |
|
|
1898 | 1671 | |
|
|
1899 | velEst = numpy.zeros((heightList.size,2))*numpy.nan | |
|
|
1900 | azimuth1 = azimuth1*numpy.pi/180 | |
|
|
1672 | #Remove Outliers | |
|
|
1673 | factor = 2 | |
|
|
1674 | wt = phaseOffset - signal.medfilt(phaseOffset,(1,5)) | |
|
|
1675 | dw = numpy.std(wt,axis = 1) | |
|
|
1676 | dw = dw.reshape((dw.size,1)) | |
|
|
1677 | ind = numpy.where(numpy.logical_or(wt>dw*factor,wt<-dw*factor)) | |
|
|
1678 | phaseOffset[ind] = numpy.nan | |
|
|
1679 | phaseOffset = stats.nanmean(phaseOffset, axis=1) | |
|
|
1901 | 1680 | |
|
|
1902 | for i in range(heightList.size): | |
|
|
1903 | h = heightList[i] | |
|
|
1904 | indH = numpy.where((metArray1[:,2] == h)&(numpy.abs(metArray1[:,-2]) < 100))[0] | |
|
|
1905 | metHeight = metArray1[indH,:] | |
|
|
1906 | if metHeight.shape[0] >= 2: | |
|
|
1907 | velAux = numpy.asmatrix(metHeight[:,-2]).T #Radial Velocities | |
|
|
1908 | iazim = metHeight[:,1].astype(int) | |
|
|
1909 | azimAux = numpy.asmatrix(azimuth1[iazim]).T #Azimuths | |
|
|
1910 | A = numpy.hstack((numpy.cos(azimAux),numpy.sin(azimAux))) | |
|
|
1911 | A = numpy.asmatrix(A) | |
|
|
1912 | A1 = numpy.linalg.pinv(A.transpose()*A)*A.transpose() | |
|
|
1913 | velHor = numpy.dot(A1,velAux) | |
|
|
1914 | ||
|
|
1915 | velEst[i,:] = numpy.squeeze(velHor) | |
|
|
1916 | return velEst | |
|
|
1681 | return phaseOffset | |
|
|
1917 | 1682 | |
|
|
1918 |
def __ |
|
|
|
1919 | meteorList = [] | |
|
|
1920 | #utctime sec1 height SNR velRad ph0 ph1 ph2 coh0 coh1 coh2 | |
|
|
1921 | #Putting back together the meteor matrix | |
|
|
1922 | utctime = metArray[:,0] | |
|
|
1923 | uniqueTime = numpy.unique(utctime) | |
|
|
1924 | ||
|
|
1925 | phaseDerThresh = 0.5 | |
|
|
1926 | ippSeconds = timeList[1] - timeList[0] | |
|
|
1927 | sec = numpy.where(timeList>1)[0][0] | |
|
|
1928 | nPairs = metArray.shape[1] - 6 | |
|
|
1929 | nHeights = len(heightList) | |
|
|
1683 | def __shiftPhase(self, data, phaseShift): | |
|
|
1684 | #this will shift the phase of a complex number | |
|
|
1685 | dataShifted = numpy.abs(data) * numpy.exp((numpy.angle(data)+phaseShift)*1j) | |
|
|
1686 | return dataShifted | |
|
|
1687 | ||
|
|
1688 | def __estimatePhaseDifference(self, array, pairslist): | |
|
|
1689 | nChannel = array.shape[0] | |
|
|
1690 | nHeights = array.shape[2] | |
|
|
1691 | numPairs = len(pairslist) | |
|
|
1692 | # phaseCCF = numpy.zeros((nChannel, 5, nHeights)) | |
|
|
1693 | phaseCCF = numpy.angle(self.__calculateCCF(array, pairslist, [-2,-1,0,1,2])) | |
|
|
1930 | 1694 | |
|
|
1931 | for t in uniqueTime: | |
|
|
1932 | metArray1 = metArray[utctime==t,:] | |
|
|
1933 | # phaseDerThresh = numpy.pi/4 #reducir Phase thresh | |
|
|
1934 | tmet = metArray1[:,1].astype(int) | |
|
|
1935 | hmet = metArray1[:,2].astype(int) | |
|
|
1936 | ||
|
|
1937 | metPhase = numpy.zeros((nPairs, heightList.size, timeList.size - 1)) | |
|
|
1938 | metPhase[:,:] = numpy.nan | |
|
|
1939 | metPhase[:,hmet,tmet] = metArray1[:,6:].T | |
|
|
1940 | ||
|
|
1941 | #Delete short trails | |
|
|
1942 | metBool = ~numpy.isnan(metPhase[0,:,:]) | |
|
|
1943 | heightVect = numpy.sum(metBool, axis = 1) | |
|
|
1944 | metBool[heightVect<sec,:] = False | |
|
|
1945 | metPhase[:,heightVect<sec,:] = numpy.nan | |
|
|
1946 | ||
|
|
1947 | #Derivative | |
|
|
1948 | metDer = numpy.abs(metPhase[:,:,1:] - metPhase[:,:,:-1]) | |
|
|
1949 | phDerAux = numpy.dstack((numpy.full((nPairs,nHeights,1), False, dtype=bool),metDer > phaseDerThresh)) | |
|
|
1950 | metPhase[phDerAux] = numpy.nan | |
|
|
1951 | ||
|
|
1952 | #--------------------------METEOR DETECTION ----------------------------------------- | |
|
|
1953 | indMet = numpy.where(numpy.any(metBool,axis=1))[0] | |
|
|
1954 | ||
|
|
1955 | for p in numpy.arange(nPairs): | |
|
|
1956 | phase = metPhase[p,:,:] | |
|
|
1957 | phDer = metDer[p,:,:] | |
|
|
1958 | ||
|
|
1959 | for h in indMet: | |
|
|
1960 | height = heightList[h] | |
|
|
1961 | phase1 = phase[h,:] #82 | |
|
|
1962 | phDer1 = phDer[h,:] | |
|
|
1963 | ||
|
|
1964 | phase1[~numpy.isnan(phase1)] = numpy.unwrap(phase1[~numpy.isnan(phase1)]) #Unwrap | |
|
|
1965 | ||
|
|
1966 | indValid = numpy.where(~numpy.isnan(phase1))[0] | |
|
|
1967 | initMet = indValid[0] | |
|
|
1968 | endMet = 0 | |
|
|
1969 | ||
|
|
1970 | for i in range(len(indValid)-1): | |
|
|
1971 | ||
|
|
1972 | #Time difference | |
|
|
1973 | inow = indValid[i] | |
|
|
1974 | inext = indValid[i+1] | |
|
|
1975 | idiff = inext - inow | |
|
|
1976 | #Phase difference | |
|
|
1977 | phDiff = numpy.abs(phase1[inext] - phase1[inow]) | |
|
|
1978 | ||
|
|
1979 | if idiff>sec or phDiff>numpy.pi/4 or inext==indValid[-1]: #End of Meteor | |
|
|
1980 | sizeTrail = inow - initMet + 1 | |
|
|
1981 | if sizeTrail>3*sec: #Too short meteors | |
|
|
1982 | x = numpy.arange(initMet,inow+1)*ippSeconds | |
|
|
1983 | y = phase1[initMet:inow+1] | |
|
|
1984 | ynnan = ~numpy.isnan(y) | |
|
|
1985 | x = x[ynnan] | |
|
|
1986 | y = y[ynnan] | |
|
|
1987 | slope, intercept, r_value, p_value, std_err = stats.linregress(x,y) | |
|
|
1988 | ylin = x*slope + intercept | |
|
|
1989 | rsq = r_value**2 | |
|
|
1990 | if rsq > 0.5: | |
|
|
1991 | vel = slope#*height*1000/(k*d) | |
|
|
1992 | estAux = numpy.array([utctime,p,height, vel, rsq]) | |
|
|
1993 | meteorList.append(estAux) | |
|
|
1994 | initMet = inext | |
|
|
1995 | metArray2 = numpy.array(meteorList) | |
|
|
1695 | #Correct phases | |
|
|
1696 | derPhaseCCF = phaseCCF[:,1:,:] - phaseCCF[:,0:-1,:] | |
|
|
1697 | indDer = numpy.where(numpy.abs(derPhaseCCF) > numpy.pi) | |
|
|
1996 | 1698 | |
|
|
1997 | return metArray2 | |
|
|
1699 | if indDer[0].shape[0] > 0: | |
|
|
1700 | for i in range(indDer[0].shape[0]): | |
|
|
1701 | signo = -numpy.sign(derPhaseCCF[indDer[0][i],indDer[1][i],indDer[2][i]]) | |
|
|
1702 | phaseCCF[indDer[0][i],indDer[1][i]+1:,:] += signo*2*numpy.pi | |
|
|
1998 | 1703 | |
|
|
1999 | def __calculateAzimuth1(self, rx_location, pairslist, azimuth0): | |
|
|
1704 | # for j in range(numSides): | |
|
|
1705 | # phaseCCFAux = self.calculateCCF(arrayCenter, arraySides[j,:,:], [-2,1,0,1,2]) | |
|
|
1706 | # phaseCCF[j,:,:] = numpy.angle(phaseCCFAux) | |
|
|
1707 | # | |
|
|
1708 | #Linear | |
|
|
1709 | phaseInt = numpy.zeros((numPairs,1)) | |
|
|
1710 | angAllCCF = phaseCCF[:,[0,1,3,4],0] | |
|
|
1711 | for j in range(numPairs): | |
|
|
1712 | fit = stats.linregress([-2,-1,1,2],angAllCCF[j,:]) | |
|
|
1713 | phaseInt[j] = fit[1] | |
|
|
1714 | #Phase Differences | |
|
|
1715 | phaseDiff = phaseInt - phaseCCF[:,2,:] | |
|
|
1716 | phaseArrival = phaseInt.reshape(phaseInt.size) | |
|
|
2000 | 1717 | |
|
|
2001 | azimuth1 = numpy.zeros(len(pairslist)) | |
|
|
2002 | dist = numpy.zeros(len(pairslist)) | |
|
|
1718 | #Dealias | |
|
|
1719 | phaseArrival = numpy.angle(numpy.exp(1j*phaseArrival)) | |
|
|
1720 | # indAlias = numpy.where(phaseArrival > numpy.pi) | |
|
|
1721 | # phaseArrival[indAlias] -= 2*numpy.pi | |
|
|
1722 | # indAlias = numpy.where(phaseArrival < -numpy.pi) | |
|
|
1723 | # phaseArrival[indAlias] += 2*numpy.pi | |
|
|
2003 | 1724 | |
|
|
2004 | for i in range(len(rx_location)): | |
|
|
2005 | ch0 = pairslist[i][0] | |
|
|
2006 | ch1 = pairslist[i][1] | |
|
|
1725 | return phaseDiff, phaseArrival | |
|
|
1726 | ||
|
|
1727 | def __coherentDetection(self, volts, timeSegment, timeInterval, pairslist, thresh): | |
|
|
1728 | #this function will run the coherent detection used in Holdworth et al. 2004 and return the net power | |
|
|
1729 | #find the phase shifts of each channel over 1 second intervals | |
|
|
1730 | #only look at ranges below the beacon signal | |
|
|
1731 | numProfPerBlock = numpy.ceil(timeSegment/timeInterval) | |
|
|
1732 | numBlocks = int(volts.shape[1]/numProfPerBlock) | |
|
|
1733 | numHeights = volts.shape[2] | |
|
|
1734 | nChannel = volts.shape[0] | |
|
|
1735 | voltsCohDet = volts.copy() | |
|
|
1736 | ||
|
|
1737 | pairsarray = numpy.array(pairslist) | |
|
|
1738 | indSides = pairsarray[:,1] | |
|
|
1739 | # indSides = numpy.array(range(nChannel)) | |
|
|
1740 | # indSides = numpy.delete(indSides, indCenter) | |
|
|
1741 | # | |
|
|
1742 | # listCenter = numpy.array_split(volts[indCenter,:,:], numBlocks, 0) | |
|
|
1743 | listBlocks = numpy.array_split(volts, numBlocks, 1) | |
|
|
1744 | ||
|
|
1745 | startInd = 0 | |
|
|
1746 | endInd = 0 | |
|
|
2007 | 1747 | |
|
|
2008 | diffX = rx_location[ch0][0] - rx_location[ch1][0] | |
|
|
2009 | diffY = rx_location[ch0][1] - rx_location[ch1][1] | |
|
|
2010 | azimuth1[i] = numpy.arctan2(diffY,diffX)*180/numpy.pi | |
|
|
2011 | dist[i] = numpy.sqrt(diffX**2 + diffY**2) | |
|
|
1748 | for i in range(numBlocks): | |
|
|
1749 | startInd = endInd | |
|
|
1750 | endInd = endInd + listBlocks[i].shape[1] | |
|
|
1751 | ||
|
|
1752 | arrayBlock = listBlocks[i] | |
|
|
1753 | # arrayBlockCenter = listCenter[i] | |
|
|
1754 | ||
|
|
1755 | #Estimate the Phase Difference | |
|
|
1756 | phaseDiff, aux = self.__estimatePhaseDifference(arrayBlock, pairslist) | |
|
|
1757 | #Phase Difference RMS | |
|
|
1758 | arrayPhaseRMS = numpy.abs(phaseDiff) | |
|
|
1759 | phaseRMSaux = numpy.sum(arrayPhaseRMS < thresh,0) | |
|
|
1760 | indPhase = numpy.where(phaseRMSaux==4) | |
|
|
1761 | #Shifting | |
|
|
1762 | if indPhase[0].shape[0] > 0: | |
|
|
1763 | for j in range(indSides.size): | |
|
|
1764 | arrayBlock[indSides[j],:,indPhase] = self.__shiftPhase(arrayBlock[indSides[j],:,indPhase], phaseDiff[j,indPhase].transpose()) | |
|
|
1765 | voltsCohDet[:,startInd:endInd,:] = arrayBlock | |
|
|
1766 | ||
|
|
1767 | return voltsCohDet | |
|
|
1768 | ||
|
|
1769 | def __calculateCCF(self, volts, pairslist ,laglist): | |
|
|
2012 | 1770 | |
|
|
2013 | azimuth1 -= azimuth0 | |
|
|
2014 | return azimuth1, dist | |
|
|
1771 | nHeights = volts.shape[2] | |
|
|
1772 | nPoints = volts.shape[1] | |
|
|
1773 | voltsCCF = numpy.zeros((len(pairslist), len(laglist), nHeights),dtype = 'complex') | |
|
|
1774 | ||
|
|
1775 | for i in range(len(pairslist)): | |
|
|
1776 | volts1 = volts[pairslist[i][0]] | |
|
|
1777 | volts2 = volts[pairslist[i][1]] | |
|
|
1778 | ||
|
|
1779 | for t in range(len(laglist)): | |
|
|
1780 | idxT = laglist[t] | |
|
|
1781 | if idxT >= 0: | |
|
|
1782 | vStacked = numpy.vstack((volts2[idxT:,:], | |
|
|
1783 | numpy.zeros((idxT, nHeights),dtype='complex'))) | |
|
|
1784 | else: | |
|
|
1785 | vStacked = numpy.vstack((numpy.zeros((-idxT, nHeights),dtype='complex'), | |
|
|
1786 | volts2[:(nPoints + idxT),:])) | |
|
|
1787 | voltsCCF[i,t,:] = numpy.sum((numpy.conjugate(volts1)*vStacked),axis=0) | |
|
|
2015 | 1788 | |
|
|
2016 | def techniqueNSM_DBS(self, **kwargs): | |
|
|
2017 | metArray = kwargs['metArray'] | |
|
|
2018 | heightList = kwargs['heightList'] | |
|
|
2019 | timeList = kwargs['timeList'] | |
|
|
2020 | zenithList = kwargs['zenithList'] | |
|
|
2021 | nChan = numpy.max(cmet) + 1 | |
|
|
2022 | nHeights = len(heightList) | |
|
|
1789 | vStacked = None | |
|
|
1790 | return voltsCCF | |
|
|
2023 | 1791 | |
|
|
2024 | utctime = metArray[:,0] | |
|
|
2025 | cmet = metArray[:,1] | |
|
|
2026 | hmet = metArray1[:,3].astype(int) | |
|
|
2027 | h1met = heightList[hmet]*zenithList[cmet] | |
|
|
2028 | vmet = metArray1[:,5] | |
|
|
1792 | def __getNoise(self, power, timeSegment, timeInterval): | |
|
|
1793 | numProfPerBlock = numpy.ceil(timeSegment/timeInterval) | |
|
|
1794 | numBlocks = int(power.shape[0]/numProfPerBlock) | |
|
|
1795 | numHeights = power.shape[1] | |
|
|
1796 | ||
|
|
1797 | listPower = numpy.array_split(power, numBlocks, 0) | |
|
|
1798 | noise = numpy.zeros((power.shape[0], power.shape[1])) | |
|
|
1799 | noise1 = numpy.zeros((power.shape[0], power.shape[1])) | |
|
|
2029 | 1800 | |
|
|
2030 | for i in range(nHeights - 1): | |
|
|
2031 | hmin = heightList[i] | |
|
|
2032 | hmax = heightList[i + 1] | |
|
|
1801 | startInd = 0 | |
|
|
1802 | endInd = 0 | |
|
|
1803 | ||
|
|
1804 | for i in range(numBlocks): #split por canal | |
|
|
1805 | startInd = endInd | |
|
|
1806 | endInd = endInd + listPower[i].shape[0] | |
|
|
2033 | 1807 | |
|
|
2034 | vthisH = vmet[(h1met>=hmin) & (h1met<hmax)] | |
|
|
1808 | arrayBlock = listPower[i] | |
|
|
1809 | noiseAux = numpy.mean(arrayBlock, 0) | |
|
|
1810 | # noiseAux = numpy.median(noiseAux) | |
|
|
1811 | # noiseAux = numpy.mean(arrayBlock) | |
|
|
1812 | noise[startInd:endInd,:] = noise[startInd:endInd,:] + noiseAux | |
|
|
2035 | 1813 | |
|
|
1814 | noiseAux1 = numpy.mean(arrayBlock) | |
|
|
1815 | noise1[startInd:endInd,:] = noise1[startInd:endInd,:] + noiseAux1 | |
|
|
1816 | ||
|
|
1817 | return noise, noise1 | |
|
|
1818 | ||
|
|
1819 | def __findMeteors(self, power, thresh): | |
|
|
1820 | nProf = power.shape[0] | |
|
|
1821 | nHeights = power.shape[1] | |
|
|
1822 | listMeteors = [] | |
|
|
1823 | ||
|
|
1824 | for i in range(nHeights): | |
|
|
1825 | powerAux = power[:,i] | |
|
|
1826 | threshAux = thresh[:,i] | |
|
|
2036 | 1827 | |
|
|
1828 | indUPthresh = numpy.where(powerAux > threshAux)[0] | |
|
|
1829 | indDNthresh = numpy.where(powerAux <= threshAux)[0] | |
|
|
2037 | 1830 | |
|
|
2038 | return data_output | |
|
|
1831 | j = 0 | |
|
|
2039 | 1832 | |
|
|
2040 | def run(self, dataOut, technique, **kwargs): | |
|
|
1833 | while (j < indUPthresh.size - 2): | |
|
|
1834 | if (indUPthresh[j + 2] == indUPthresh[j] + 2): | |
|
|
1835 | indDNAux = numpy.where(indDNthresh > indUPthresh[j]) | |
|
|
1836 | indDNthresh = indDNthresh[indDNAux] | |
|
|
1837 | ||
|
|
1838 | if (indDNthresh.size > 0): | |
|
|
1839 | indEnd = indDNthresh[0] - 1 | |
|
|
1840 | indInit = indUPthresh[j] | |
|
|
1841 | ||
|
|
1842 | meteor = powerAux[indInit:indEnd + 1] | |
|
|
1843 | indPeak = meteor.argmax() + indInit | |
|
|
1844 | FLA = sum(numpy.conj(meteor)*numpy.hstack((meteor[1:],0))) | |
|
|
1845 | ||
|
|
1846 | listMeteors.append(numpy.array([i,indInit,indPeak,indEnd,FLA])) #CHEQUEAR!!!!! | |
|
|
1847 | j = numpy.where(indUPthresh == indEnd)[0] + 1 | |
|
|
1848 | else: j+=1 | |
|
|
1849 | else: j+=1 | |
|
|
1850 | ||
|
|
1851 | return listMeteors | |
|
|
1852 | ||
|
|
1853 | def __removeMultipleDetections(self,listMeteors, rangeLimit, timeLimit): | |
|
|
2041 | 1854 | |
|
|
2042 | param = dataOut.data_param | |
|
|
2043 | if dataOut.abscissaList != None: | |
|
|
2044 | absc = dataOut.abscissaList[:-1] | |
|
|
2045 | noise = dataOut.noise | |
|
|
2046 | heightList = dataOut.heightList | |
|
|
2047 | SNR = dataOut.data_SNR | |
|
|
1855 | arrayMeteors = numpy.asarray(listMeteors) | |
|
|
1856 | listMeteors1 = [] | |
|
|
2048 | 1857 | |
|
|
2049 | if technique == 'DBS': | |
|
|
1858 | while arrayMeteors.shape[0] > 0: | |
|
|
1859 | FLAs = arrayMeteors[:,4] | |
|
|
1860 | maxFLA = FLAs.argmax() | |
|
|
1861 | listMeteors1.append(arrayMeteors[maxFLA,:]) | |
|
|
2050 | 1862 | |
|
|
2051 | if kwargs.has_key('dirCosx') and kwargs.has_key('dirCosy'): | |
|
|
2052 | theta_x = numpy.array(kwargs['dirCosx']) | |
|
|
2053 | theta_y = numpy.array(kwargs['dirCosy']) | |
|
|
2054 | else: | |
|
|
2055 | elev = numpy.array(kwargs['elevation']) | |
|
|
2056 | azim = numpy.array(kwargs['azimuth']) | |
|
|
2057 | theta_x, theta_y = self.__calculateCosDir(elev, azim) | |
|
|
2058 | azimuth = kwargs['correctAzimuth'] | |
|
|
2059 | if kwargs.has_key('horizontalOnly'): | |
|
|
2060 | horizontalOnly = kwargs['horizontalOnly'] | |
|
|
2061 | else: horizontalOnly = False | |
|
|
2062 | if kwargs.has_key('correctFactor'): | |
|
|
2063 | correctFactor = kwargs['correctFactor'] | |
|
|
2064 | else: correctFactor = 1 | |
|
|
2065 | if kwargs.has_key('channelList'): | |
|
|
2066 | channelList = kwargs['channelList'] | |
|
|
2067 | if len(channelList) == 2: | |
|
|
2068 | horizontalOnly = True | |
|
|
2069 | arrayChannel = numpy.array(channelList) | |
|
|
2070 | param = param[arrayChannel,:,:] | |
|
|
2071 | theta_x = theta_x[arrayChannel] | |
|
|
2072 | theta_y = theta_y[arrayChannel] | |
|
|
1863 | MeteorInitTime = arrayMeteors[maxFLA,1] | |
|
|
1864 | MeteorEndTime = arrayMeteors[maxFLA,3] | |
|
|
1865 | MeteorHeight = arrayMeteors[maxFLA,0] | |
|
|
2073 | 1866 | |
|
|
2074 | velRadial0 = param[:,1,:] #Radial velocity | |
|
|
2075 | dataOut.data_output, dataOut.heightList, dataOut.data_SNR = self.techniqueDBS(velRadial0, theta_x, theta_y, azimuth, correctFactor, horizontalOnly, heightList, SNR) #DBS Function | |
|
|
2076 | dataOut.utctimeInit = dataOut.utctime | |
|
|
2077 | dataOut.outputInterval = dataOut.paramInterval | |
|
|
1867 | #Check neighborhood | |
|
|
1868 | maxHeightIndex = MeteorHeight + rangeLimit | |
|
|
1869 | minHeightIndex = MeteorHeight - rangeLimit | |
|
|
1870 | minTimeIndex = MeteorInitTime - timeLimit | |
|
|
1871 | maxTimeIndex = MeteorEndTime + timeLimit | |
|
|
2078 | 1872 | |
|
|
2079 | elif technique == 'SA': | |
|
|
1873 | #Check Heights | |
|
|
1874 | indHeight = numpy.logical_and(arrayMeteors[:,0] >= minHeightIndex, arrayMeteors[:,0] <= maxHeightIndex) | |
|
|
1875 | indTime = numpy.logical_and(arrayMeteors[:,3] >= minTimeIndex, arrayMeteors[:,1] <= maxTimeIndex) | |
|
|
1876 | indBoth = numpy.where(numpy.logical_and(indTime,indHeight)) | |
|
|
1877 | ||
|
|
1878 | arrayMeteors = numpy.delete(arrayMeteors, indBoth, axis = 0) | |
|
|
2080 | 1879 | |
|
|
2081 | #Parameters | |
|
|
2082 | position_x = kwargs['positionX'] | |
|
|
2083 | position_y = kwargs['positionY'] | |
|
|
2084 | azimuth = kwargs['azimuth'] | |
|
|
1880 | return listMeteors1 | |
|
|
1881 | ||
|
|
1882 | def __meteorReestimation(self, listMeteors, volts, pairslist, thresh, noise, timeInterval,frequency): | |
|
|
1883 | numHeights = volts.shape[2] | |
|
|
1884 | nChannel = volts.shape[0] | |
|
|
1885 | ||
|
|
1886 | thresholdPhase = thresh[0] | |
|
|
1887 | thresholdNoise = thresh[1] | |
|
|
1888 | thresholdDB = float(thresh[2]) | |
|
|
1889 | ||
|
|
1890 | thresholdDB1 = 10**(thresholdDB/10) | |
|
|
1891 | pairsarray = numpy.array(pairslist) | |
|
|
1892 | indSides = pairsarray[:,1] | |
|
|
1893 | ||
|
|
1894 | pairslist1 = list(pairslist) | |
|
|
1895 | pairslist1.append((0,1)) | |
|
|
1896 | pairslist1.append((3,4)) | |
|
|
1897 | ||
|
|
1898 | listMeteors1 = [] | |
|
|
1899 | listPowerSeries = [] | |
|
|
1900 | listVoltageSeries = [] | |
|
|
1901 | #volts has the war data | |
|
|
1902 | ||
|
|
1903 | if frequency == 30e6: | |
|
|
1904 | timeLag = 45*10**-3 | |
|
|
1905 | else: | |
|
|
1906 | timeLag = 15*10**-3 | |
|
|
1907 | lag = numpy.ceil(timeLag/timeInterval) | |
|
|
1908 | ||
|
|
1909 | for i in range(len(listMeteors)): | |
|
|
1910 | ||
|
|
1911 | ###################### 3.6 - 3.7 PARAMETERS REESTIMATION ######################### | |
|
|
1912 | meteorAux = numpy.zeros(16) | |
|
|
1913 | ||
|
|
1914 | #Loading meteor Data (mHeight, mStart, mPeak, mEnd) | |
|
|
1915 | mHeight = listMeteors[i][0] | |
|
|
1916 | mStart = listMeteors[i][1] | |
|
|
1917 | mPeak = listMeteors[i][2] | |
|
|
1918 | mEnd = listMeteors[i][3] | |
|
|
1919 | ||
|
|
1920 | #get the volt data between the start and end times of the meteor | |
|
|
1921 | meteorVolts = volts[:,mStart:mEnd+1,mHeight] | |
|
|
1922 | meteorVolts = meteorVolts.reshape(meteorVolts.shape[0], meteorVolts.shape[1], 1) | |
|
|
2085 | 1923 | |
|
|
2086 | if kwargs.has_key('crosspairsList'): | |
|
|
2087 | pairs = kwargs['crosspairsList'] | |
|
|
2088 | else: | |
|
|
2089 | pairs = None | |
|
|
2090 | ||
|
|
2091 | if kwargs.has_key('correctFactor'): | |
|
|
2092 | correctFactor = kwargs['correctFactor'] | |
|
|
2093 | else: | |
|
|
2094 | correctFactor = 1 | |
|
|
2095 | ||
|
|
2096 | tau = dataOut.data_param | |
|
|
2097 | _lambda = dataOut.C/dataOut.frequency | |
|
|
2098 | pairsList = dataOut.groupList | |
|
|
2099 | nChannels = dataOut.nChannels | |
|
|
2100 | ||
|
|
2101 | dataOut.data_output = self.techniqueSA(pairs, pairsList, nChannels, tau, azimuth, _lambda, position_x, position_y, absc, correctFactor) | |
|
|
2102 | dataOut.utctimeInit = dataOut.utctime | |
|
|
2103 | dataOut.outputInterval = dataOut.timeInterval | |
|
|
1924 | #3.6. Phase Difference estimation | |
|
|
1925 | phaseDiff, aux = self.__estimatePhaseDifference(meteorVolts, pairslist) | |
|
|
2104 | 1926 | |
|
|
2105 | elif technique == 'Meteors': | |
|
|
2106 | dataOut.flagNoData = True | |
|
|
2107 | self.__dataReady = False | |
|
|
1927 | #3.7. Phase difference removal & meteor start, peak and end times reestimated | |
|
|
1928 | #meteorVolts0.- all Channels, all Profiles | |
|
|
1929 | meteorVolts0 = volts[:,:,mHeight] | |
|
|
1930 | meteorThresh = noise[:,mHeight]*thresholdNoise | |
|
|
1931 | meteorNoise = noise[:,mHeight] | |
|
|
1932 | meteorVolts0[indSides,:] = self.__shiftPhase(meteorVolts0[indSides,:], phaseDiff) #Phase Shifting | |
|
|
1933 | powerNet0 = numpy.nansum(numpy.abs(meteorVolts0)**2, axis = 0) #Power | |
|
|
2108 | 1934 | |
|
|
2109 | if kwargs.has_key('nHours'): | |
|
|
2110 | nHours = kwargs['nHours'] | |
|
|
2111 | else: | |
|
|
2112 |
|
|
|
|
2113 | ||
|
|
2114 | if kwargs.has_key('meteorsPerBin'): | |
|
|
2115 | meteorThresh = kwargs['meteorsPerBin'] | |
|
|
2116 | else: | |
|
|
2117 | meteorThresh = 6 | |
|
|
1935 | #Times reestimation | |
|
|
1936 | mStart1 = numpy.where(powerNet0[:mPeak] < meteorThresh[:mPeak])[0] | |
|
|
1937 | if mStart1.size > 0: | |
|
|
1938 | mStart1 = mStart1[-1] + 1 | |
|
|
2118 | 1939 | |
|
|
2119 | if kwargs.has_key('hmin'): | |
|
|
2120 |
|
|
|
|
2121 | else: hmin = 70 | |
|
|
2122 | if kwargs.has_key('hmax'): | |
|
|
2123 | hmax = kwargs['hmax'] | |
|
|
2124 | else: hmax = 110 | |
|
|
2125 | ||
|
|
2126 | dataOut.outputInterval = nHours*3600 | |
|
|
2127 | ||
|
|
2128 | if self.__isConfig == False: | |
|
|
2129 | # self.__initime = dataOut.datatime.replace(minute = 0, second = 0, microsecond = 03) | |
|
|
2130 | #Get Initial LTC time | |
|
|
2131 | self.__initime = datetime.datetime.utcfromtimestamp(dataOut.utctime) | |
|
|
2132 | self.__initime = (self.__initime.replace(minute = 0, second = 0, microsecond = 0) - datetime.datetime(1970, 1, 1)).total_seconds() | |
|
|
2133 | ||
|
|
2134 | self.__isConfig = True | |
|
|
1940 | else: | |
|
|
1941 | mStart1 = mPeak | |
|
|
2135 | 1942 | |
|
|
2136 | if self.__buffer == None: | |
|
|
2137 | self.__buffer = dataOut.data_param | |
|
|
2138 | self.__firstdata = copy.copy(dataOut) | |
|
|
2139 | ||
|
|
1943 | mEnd1 = numpy.where(powerNet0[mPeak:] < meteorThresh[mPeak:])[0][0] + mPeak - 1 | |
|
|
1944 | mEndDecayTime1 = numpy.where(powerNet0[mPeak:] < meteorNoise[mPeak:])[0] | |
|
|
1945 | if mEndDecayTime1.size == 0: | |
|
|
1946 | mEndDecayTime1 = powerNet0.size | |
|
|
2140 | 1947 | else: |
|
|
2141 | self.__buffer = numpy.vstack((self.__buffer, dataOut.data_param)) | |
|
|
1948 | mEndDecayTime1 = mEndDecayTime1[0] + mPeak - 1 | |
|
|
1949 | # mPeak1 = meteorVolts0[mStart1:mEnd1 + 1].argmax() | |
|
|
2142 | 1950 | |
|
|
2143 | self.__checkTime(dataOut.utctime, dataOut.paramInterval, dataOut.outputInterval) #Check if the buffer is ready | |
|
|
1951 | #meteorVolts1.- all Channels, from start to end | |
|
|
1952 | meteorVolts1 = meteorVolts0[:,mStart1:mEnd1 + 1] | |
|
|
1953 | meteorVolts2 = meteorVolts0[:,mPeak + lag:mEnd1 + 1] | |
|
|
1954 | if meteorVolts2.shape[1] == 0: | |
|
|
1955 | meteorVolts2 = meteorVolts0[:,mPeak:mEnd1 + 1] | |
|
|
1956 | meteorVolts1 = meteorVolts1.reshape(meteorVolts1.shape[0], meteorVolts1.shape[1], 1) | |
|
|
1957 | meteorVolts2 = meteorVolts2.reshape(meteorVolts2.shape[0], meteorVolts2.shape[1], 1) | |
|
|
1958 | ##################### END PARAMETERS REESTIMATION ######################### | |
|
|
2144 | 1959 | |
|
|
2145 | if self.__dataReady: | |
|
|
2146 | dataOut.utctimeInit = self.__initime | |
|
|
1960 | ##################### 3.8 PHASE DIFFERENCE REESTIMATION ######################## | |
|
|
1961 | # if mEnd1 - mStart1 > 4: #Error Number 6: echo less than 5 samples long; too short for analysis | |
|
|
1962 | if meteorVolts2.shape[1] > 0: | |
|
|
1963 | #Phase Difference re-estimation | |
|
|
1964 | phaseDiff1, phaseDiffint = self.__estimatePhaseDifference(meteorVolts2, pairslist1) #Phase Difference Estimation | |
|
|
1965 | # phaseDiff1, phaseDiffint = self.estimatePhaseDifference(meteorVolts2, pairslist) | |
|
|
1966 | meteorVolts2 = meteorVolts2.reshape(meteorVolts2.shape[0], meteorVolts2.shape[1]) | |
|
|
1967 | phaseDiff11 = numpy.reshape(phaseDiff1, (phaseDiff1.shape[0],1)) | |
|
|
1968 | meteorVolts2[indSides,:] = self.__shiftPhase(meteorVolts2[indSides,:], phaseDiff11[0:4]) #Phase Shifting | |
|
|
2147 | 1969 | |
|
|
2148 | self.__initime += dataOut.outputInterval #to erase time offset | |
|
|
1970 | #Phase Difference RMS | |
|
|
1971 | phaseRMS1 = numpy.sqrt(numpy.mean(numpy.square(phaseDiff1))) | |
|
|
1972 | powerNet1 = numpy.nansum(numpy.abs(meteorVolts1[:,:])**2,0) | |
|
|
1973 | #Data from Meteor | |
|
|
1974 | mPeak1 = powerNet1.argmax() + mStart1 | |
|
|
1975 | mPeakPower1 = powerNet1.max() | |
|
|
1976 | noiseAux = sum(noise[mStart1:mEnd1 + 1,mHeight]) | |
|
|
1977 | mSNR1 = (sum(powerNet1)-noiseAux)/noiseAux | |
|
|
1978 | Meteor1 = numpy.array([mHeight, mStart1, mPeak1, mEnd1, mPeakPower1, mSNR1, phaseRMS1]) | |
|
|
1979 | Meteor1 = numpy.hstack((Meteor1,phaseDiffint)) | |
|
|
1980 | PowerSeries = powerNet0[mStart1:mEndDecayTime1 + 1] | |
|
|
1981 | #Vectorize | |
|
|
1982 | meteorAux[0:7] = [mHeight, mStart1, mPeak1, mEnd1, mPeakPower1, mSNR1, phaseRMS1] | |
|
|
1983 | meteorAux[7:11] = phaseDiffint[0:4] | |
|
|
2149 | 1984 | |
|
|
2150 | dataOut.data_output, dataOut.heightList = self.techniqueMeteors(self.__buffer, meteorThresh, hmin, hmax) | |
|
|
2151 | dataOut.flagNoData = False | |
|
|
2152 | self.__buffer = None | |
|
|
1985 | #Rejection Criterions | |
|
|
1986 | if phaseRMS1 > thresholdPhase: #Error Number 17: Phase variation | |
|
|
1987 | meteorAux[-1] = 17 | |
|
|
1988 | elif mSNR1 < thresholdDB1: #Error Number 1: SNR < threshold dB | |
|
|
1989 | meteorAux[-1] = 1 | |
|
|
2153 | 1990 | |
|
|
2154 | elif technique == 'Meteors1': | |
|
|
2155 | dataOut.flagNoData = True | |
|
|
2156 | self.__dataReady = False | |
|
|
2157 | ||
|
|
2158 | if kwargs.has_key('nMins'): | |
|
|
2159 | nMins = kwargs['nMins'] | |
|
|
2160 | else: nMins = 20 | |
|
|
2161 | if kwargs.has_key('rx_location'): | |
|
|
2162 | rx_location = kwargs['rx_location'] | |
|
|
2163 | else: rx_location = [(0,1),(1,1),(1,0)] | |
|
|
2164 | if kwargs.has_key('azimuth'): | |
|
|
2165 | azimuth = kwargs['azimuth'] | |
|
|
2166 | else: azimuth = 51 | |
|
|
2167 | if kwargs.has_key('dfactor'): | |
|
|
2168 | dfactor = kwargs['dfactor'] | |
|
|
2169 | if kwargs.has_key('mode'): | |
|
|
2170 | mode = kwargs['mode'] | |
|
|
2171 | else: mode = 'SA' | |
|
|
2172 | 1991 | |
|
|
2173 | #Borrar luego esto | |
|
|
2174 | if dataOut.groupList == None: | |
|
|
2175 | dataOut.groupList = [(0,1),(0,2),(1,2)] | |
|
|
2176 | groupList = dataOut.groupList | |
|
|
2177 |
|
|
|
|
2178 | freq = 50e6 | |
|
|
2179 | lamb = C/freq | |
|
|
2180 | k = 2*numpy.pi/lamb | |
|
|
1992 | else: | |
|
|
1993 | meteorAux[0:4] = [mHeight, mStart, mPeak, mEnd] | |
|
|
1994 | meteorAux[-1] = 6 #Error Number 6: echo less than 5 samples long; too short for analysis | |
|
|
1995 | PowerSeries = 0 | |
|
|
1996 | ||
|
|
1997 | listMeteors1.append(meteorAux) | |
|
|
1998 | listPowerSeries.append(PowerSeries) | |
|
|
1999 | listVoltageSeries.append(meteorVolts1) | |
|
|
2000 | ||
|
|
2001 | return listMeteors1, listPowerSeries, listVoltageSeries | |
|
|
2002 | ||
|
|
2003 | def __estimateDecayTime(self, listMeteors, listPower, timeInterval, frequency): | |
|
|
2004 | ||
|
|
2005 | threshError = 10 | |
|
|
2006 | #Depending if it is 30 or 50 MHz | |
|
|
2007 | if frequency == 30e6: | |
|
|
2008 | timeLag = 45*10**-3 | |
|
|
2009 | else: | |
|
|
2010 | timeLag = 15*10**-3 | |
|
|
2011 | lag = numpy.ceil(timeLag/timeInterval) | |
|
|
2012 | ||
|
|
2013 | listMeteors1 = [] | |
|
|
2014 | ||
|
|
2015 | for i in range(len(listMeteors)): | |
|
|
2016 | meteorPower = listPower[i] | |
|
|
2017 | meteorAux = listMeteors[i] | |
|
|
2181 | 2018 | |
|
|
2182 | timeList = dataOut.abscissaList | |
|
|
2183 | heightList = dataOut.heightList | |
|
|
2019 | if meteorAux[-1] == 0: | |
|
|
2020 | ||
|
|
2021 | try: | |
|
|
2022 | indmax = meteorPower.argmax() | |
|
|
2023 | indlag = indmax + lag | |
|
|
2024 | ||
|
|
2025 | y = meteorPower[indlag:] | |
|
|
2026 | x = numpy.arange(0, y.size)*timeLag | |
|
|
2027 | ||
|
|
2028 | #first guess | |
|
|
2029 | a = y[0] | |
|
|
2030 | tau = timeLag | |
|
|
2031 | #exponential fit | |
|
|
2032 | popt, pcov = optimize.curve_fit(self.__exponential_function, x, y, p0 = [a, tau]) | |
|
|
2033 | y1 = self.__exponential_function(x, *popt) | |
|
|
2034 | #error estimation | |
|
|
2035 | error = sum((y - y1)**2)/(numpy.var(y)*(y.size - popt.size)) | |
|
|
2036 | ||
|
|
2037 | decayTime = popt[1] | |
|
|
2038 | riseTime = indmax*timeInterval | |
|
|
2039 | meteorAux[11:13] = [decayTime, error] | |
|
|
2040 | ||
|
|
2041 | #Table items 7, 8 and 11 | |
|
|
2042 | if (riseTime > 0.3): #Number 7: Echo rise exceeds 0.3s | |
|
|
2043 | meteorAux[-1] = 7 | |
|
|
2044 | elif (decayTime < 2*riseTime) : #Number 8: Echo decay time less than than twice rise time | |
|
|
2045 | meteorAux[-1] = 8 | |
|
|
2046 | if (error > threshError): #Number 11: Poor fit to amplitude for estimation of decay time | |
|
|
2047 | meteorAux[-1] = 11 | |
|
|
2048 | ||
|
|
2049 | ||
|
|
2050 | except: | |
|
|
2051 | meteorAux[-1] = 11 | |
|
|
2052 | ||
|
|
2184 | 2053 | |
|
|
2185 | if self.__isConfig == False: | |
|
|
2186 | dataOut.outputInterval = nMins*60 | |
|
|
2187 | # self.__initime = dataOut.datatime.replace(minute = 0, second = 0, microsecond = 03) | |
|
|
2188 | #Get Initial LTC time | |
|
|
2189 | initime = datetime.datetime.utcfromtimestamp(dataOut.utctime) | |
|
|
2190 | minuteAux = initime.minute | |
|
|
2191 | minuteNew = int(numpy.floor(minuteAux/nMins)*nMins) | |
|
|
2192 | self.__initime = (initime.replace(minute = minuteNew, second = 0, microsecond = 0) - datetime.datetime(1970, 1, 1)).total_seconds() | |
|
|
2054 | listMeteors1.append(meteorAux) | |
|
|
2055 | ||
|
|
2056 | return listMeteors1 | |
|
|
2193 | 2057 | |
|
|
2194 | self.__isConfig = True | |
|
|
2058 | #Exponential Function | |
|
|
2059 | ||
|
|
2060 | def __exponential_function(self, x, a, tau): | |
|
|
2061 | y = a*numpy.exp(-x/tau) | |
|
|
2062 | return y | |
|
|
2063 | ||
|
|
2064 | def __getRadialVelocity(self, listMeteors, listVolts, radialStdThresh, pairslist, timeInterval): | |
|
|
2065 | ||
|
|
2066 | pairslist1 = list(pairslist) | |
|
|
2067 | pairslist1.append((0,1)) | |
|
|
2068 | pairslist1.append((3,4)) | |
|
|
2069 | numPairs = len(pairslist1) | |
|
|
2070 | #Time Lag | |
|
|
2071 | timeLag = 45*10**-3 | |
|
|
2072 | c = 3e8 | |
|
|
2073 | lag = numpy.ceil(timeLag/timeInterval) | |
|
|
2074 | freq = 30e6 | |
|
|
2075 | ||
|
|
2076 | listMeteors1 = [] | |
|
|
2077 | ||
|
|
2078 | for i in range(len(listMeteors)): | |
|
|
2079 | meteorAux = listMeteors[i] | |
|
|
2080 | if meteorAux[-1] == 0: | |
|
|
2081 | mStart = listMeteors[i][1] | |
|
|
2082 | mPeak = listMeteors[i][2] | |
|
|
2083 | mLag = mPeak - mStart + lag | |
|
|
2084 | ||
|
|
2085 | #get the volt data between the start and end times of the meteor | |
|
|
2086 | meteorVolts = listVolts[i] | |
|
|
2087 | meteorVolts = meteorVolts.reshape(meteorVolts.shape[0], meteorVolts.shape[1], 1) | |
|
|
2088 | ||
|
|
2089 | #Get CCF | |
|
|
2090 | allCCFs = self.__calculateCCF(meteorVolts, pairslist1, [-2,-1,0,1,2]) | |
|
|
2091 | ||
|
|
2092 | #Method 2 | |
|
|
2093 | slopes = numpy.zeros(numPairs) | |
|
|
2094 | time = numpy.array([-2,-1,1,2])*timeInterval | |
|
|
2095 | angAllCCF = numpy.angle(allCCFs[:,[0,1,3,4],0]) | |
|
|
2096 | ||
|
|
2097 | #Correct phases | |
|
|
2098 | derPhaseCCF = angAllCCF[:,1:] - angAllCCF[:,0:-1] | |
|
|
2099 | indDer = numpy.where(numpy.abs(derPhaseCCF) > numpy.pi) | |
|
|
2195 | 2100 | |
|
|
2196 | if self.__buffer == None: | |
|
|
2197 | self.__buffer = dataOut.data_param | |
|
|
2198 | self.__firstdata = copy.copy(dataOut) | |
|
|
2101 | if indDer[0].shape[0] > 0: | |
|
|
2102 | for i in range(indDer[0].shape[0]): | |
|
|
2103 | signo = -numpy.sign(derPhaseCCF[indDer[0][i],indDer[1][i]]) | |
|
|
2104 | angAllCCF[indDer[0][i],indDer[1][i]+1:] += signo*2*numpy.pi | |
|
|
2199 | 2105 | |
|
|
2200 | else: | |
|
|
2201 | self.__buffer = numpy.vstack((self.__buffer, dataOut.data_param)) | |
|
|
2202 | ||
|
|
2203 | self.__checkTime(dataOut.utctime, dataOut.paramInterval, dataOut.outputInterval) #Check if the buffer is ready | |
|
|
2204 | ||
|
|
2205 | if self.__dataReady: | |
|
|
2206 | dataOut.utctimeInit = self.__initime | |
|
|
2207 | self.__initime += dataOut.outputInterval #to erase time offset | |
|
|
2106 | # fit = scipy.stats.linregress(numpy.array([-2,-1,1,2])*timeInterval, numpy.array([phaseLagN2s[i],phaseLagN1s[i],phaseLag1s[i],phaseLag2s[i]])) | |
|
|
2107 | for j in range(numPairs): | |
|
|
2108 | fit = stats.linregress(time, angAllCCF[j,:]) | |
|
|
2109 | slopes[j] = fit[0] | |
|
|
2208 | 2110 | |
|
|
2209 |
|
|
|
|
2210 | if mode == 'SA': | |
|
|
2211 | dataOut.data_output = self.techniqueNSM_SA(rx_location=rx_location, groupList=groupList, azimuth=azimuth, dfactor=dfactor, k=k,metArray=metArray, heightList=heightList,timeList=timeList) | |
|
|
2212 | elif mode == 'DBS': | |
|
|
2213 | dataOut.data_output = self.techniqueNSM_DBS(metArray=metArray,heightList=heightList,timeList=timeList) | |
|
|
2214 | dataOut.data_output = dataOut.data_output.T | |
|
|
2215 | dataOut.flagNoData = False | |
|
|
2216 | self.__buffer = None | |
|
|
2217 | ||
|
|
2218 | return | |
|
|
2219 | ||
|
|
2220 | class EWDriftsEstimation(Operation): | |
|
|
2221 | ||
|
|
2222 | def __init__(self): | |
|
|
2223 | Operation.__init__(self) | |
|
|
2111 | #Remove Outlier | |
|
|
2112 | # indOut = numpy.argmax(numpy.abs(slopes - numpy.mean(slopes))) | |
|
|
2113 | # slopes = numpy.delete(slopes,indOut) | |
|
|
2114 | # indOut = numpy.argmax(numpy.abs(slopes - numpy.mean(slopes))) | |
|
|
2115 | # slopes = numpy.delete(slopes,indOut) | |
|
|
2116 | ||
|
|
2117 | radialVelocity = -numpy.mean(slopes)*(0.25/numpy.pi)*(c/freq) | |
|
|
2118 | radialError = numpy.std(slopes)*(0.25/numpy.pi)*(c/freq) | |
|
|
2119 | meteorAux[-2] = radialError | |
|
|
2120 | meteorAux[-3] = radialVelocity | |
|
|
2121 | ||
|
|
2122 | #Setting Error | |
|
|
2123 | #Number 15: Radial Drift velocity or projected horizontal velocity exceeds 200 m/s | |
|
|
2124 | if numpy.abs(radialVelocity) > 200: | |
|
|
2125 | meteorAux[-1] = 15 | |
|
|
2126 | #Number 12: Poor fit to CCF variation for estimation of radial drift velocity | |
|
|
2127 | elif radialError > radialStdThresh: | |
|
|
2128 | meteorAux[-1] = 12 | |
|
|
2129 | ||
|
|
2130 | listMeteors1.append(meteorAux) | |
|
|
2131 | return listMeteors1 | |
|
|
2224 | 2132 | |
|
|
2225 | def __correctValues(self, heiRang, phi, velRadial, SNR): | |
|
|
2226 | listPhi = phi.tolist() | |
|
|
2227 | maxid = listPhi.index(max(listPhi)) | |
|
|
2228 | minid = listPhi.index(min(listPhi)) | |
|
|
2133 | def __setNewArrays(self, listMeteors, date, heiRang): | |
|
|
2229 | 2134 | |
|
|
2230 | rango = range(len(phi)) | |
|
|
2231 | # rango = numpy.delete(rango,maxid) | |
|
|
2135 | #New arrays | |
|
|
2136 | arrayMeteors = numpy.array(listMeteors) | |
|
|
2137 | arrayParameters = numpy.zeros((len(listMeteors), 13)) | |
|
|
2232 | 2138 | |
|
|
2233 | heiRang1 = heiRang*math.cos(phi[maxid]) | |
|
|
2234 | heiRangAux = heiRang*math.cos(phi[minid]) | |
|
|
2235 | indOut = (heiRang1 < heiRangAux[0]).nonzero() | |
|
|
2236 | heiRang1 = numpy.delete(heiRang1,indOut) | |
|
|
2139 | #Date inclusion | |
|
|
2140 | # date = re.findall(r'\((.*?)\)', date) | |
|
|
2141 | # date = date[0].split(',') | |
|
|
2142 | # date = map(int, date) | |
|
|
2143 | # | |
|
|
2144 | # if len(date)<6: | |
|
|
2145 | # date.append(0) | |
|
|
2146 | # | |
|
|
2147 | # date = [date[0]*10000 + date[1]*100 + date[2], date[3]*10000 + date[4]*100 + date[5]] | |
|
|
2148 | # arrayDate = numpy.tile(date, (len(listMeteors), 1)) | |
|
|
2149 | arrayDate = numpy.tile(date, (len(listMeteors))) | |
|
|
2237 | 2150 | |
|
|
2238 | velRadial1 = numpy.zeros([len(phi),len(heiRang1)]) | |
|
|
2239 | SNR1 = numpy.zeros([len(phi),len(heiRang1)]) | |
|
|
2151 | #Meteor array | |
|
|
2152 | # arrayMeteors[:,0] = heiRang[arrayMeteors[:,0].astype(int)] | |
|
|
2153 | # arrayMeteors = numpy.hstack((arrayDate, arrayMeteors)) | |
|
|
2240 | 2154 | |
|
|
2241 | for i in rango: | |
|
|
2242 | x = heiRang*math.cos(phi[i]) | |
|
|
2243 | y1 = velRadial[i,:] | |
|
|
2244 | f1 = interpolate.interp1d(x,y1,kind = 'cubic') | |
|
|
2245 | ||
|
|
2246 | x1 = heiRang1 | |
|
|
2247 | y11 = f1(x1) | |
|
|
2248 | ||
|
|
2249 | y2 = SNR[i,:] | |
|
|
2250 | f2 = interpolate.interp1d(x,y2,kind = 'cubic') | |
|
|
2251 | y21 = f2(x1) | |
|
|
2252 | ||
|
|
2253 | velRadial1[i,:] = y11 | |
|
|
2254 | SNR1[i,:] = y21 | |
|
|
2255 | ||
|
|
2256 | return heiRang1, velRadial1, SNR1 | |
|
|
2155 | #Parameters Array | |
|
|
2156 | arrayParameters[:,0] = arrayDate #Date | |
|
|
2157 | arrayParameters[:,1] = heiRang[arrayMeteors[:,0].astype(int)] #Range | |
|
|
2158 | arrayParameters[:,6:8] = arrayMeteors[:,-3:-1] #Radial velocity and its error | |
|
|
2159 | arrayParameters[:,8:12] = arrayMeteors[:,7:11] #Phases | |
|
|
2160 | arrayParameters[:,-1] = arrayMeteors[:,-1] #Error | |
|
|
2257 | 2161 | |
|
|
2258 | def run(self, dataOut, zenith, zenithCorrection): | |
|
|
2259 | heiRang = dataOut.heightList | |
|
|
2260 | velRadial = dataOut.data_param[:,3,:] | |
|
|
2261 | SNR = dataOut.data_SNR | |
|
|
2262 | 2162 | |
|
|
2263 | zenith = numpy.array(zenith) | |
|
|
2264 | zenith -= zenithCorrection | |
|
|
2265 | zenith *= numpy.pi/180 | |
|
|
2266 |
|
|
|
|
2267 | heiRang1, velRadial1, SNR1 = self.__correctValues(heiRang, numpy.abs(zenith), velRadial, SNR) | |
|
|
2268 | ||
|
|
2269 | alp = zenith[0] | |
|
|
2270 |
|
|
|
|
2271 | ||
|
|
2272 | w_w = velRadial1[0,:] | |
|
|
2273 | w_e = velRadial1[1,:] | |
|
|
2274 | ||
|
|
2275 | w = (w_w*numpy.sin(bet) - w_e*numpy.sin(alp))/(numpy.cos(alp)*numpy.sin(bet) - numpy.cos(bet)*numpy.sin(alp)) | |
|
|
2276 | u = (w_w*numpy.cos(bet) - w_e*numpy.cos(alp))/(numpy.sin(alp)*numpy.cos(bet) - numpy.sin(bet)*numpy.cos(alp)) | |
|
|
2163 | return arrayParameters | |
|
|
2164 | ||
|
|
2165 | class CorrectMeteorPhases(Operation): | |
|
|
2166 | ||
|
|
2167 | def run(self, dataOut, phaseOffsets, hmin = 50, hmax = 150, azimuth = 45, channelPositions = None): | |
|
|
2168 | ||
|
|
2169 | arrayParameters = dataOut.data_param | |
|
|
2170 | pairsList = [] | |
|
|
2171 | pairx = (0,1) | |
|
|
2172 | pairy = (2,3) | |
|
|
2173 | pairsList.append(pairx) | |
|
|
2174 | pairsList.append(pairy) | |
|
|
2175 | jph = numpy.zeros(4) | |
|
|
2277 | 2176 | |
|
|
2278 | winds = numpy.vstack((u,w)) | |
|
|
2177 | phaseOffsets = numpy.array(phaseOffsets)*numpy.pi/180 | |
|
|
2178 | # arrayParameters[:,8:12] = numpy.unwrap(arrayParameters[:,8:12] + phaseOffsets) | |
|
|
2179 | arrayParameters[:,8:12] = numpy.angle(numpy.exp(1j*(arrayParameters[:,8:12] + phaseOffsets))) | |
|
|
2279 | 2180 | |
|
|
2280 | dataOut.heightList = heiRang1 | |
|
|
2281 | dataOut.data_output = winds | |
|
|
2282 | dataOut.data_SNR = SNR1 | |
|
|
2181 | meteorOps = MeteorOperations() | |
|
|
2182 | if channelPositions == None: | |
|
|
2183 | # channelPositions = [(2.5,0), (0,2.5), (0,0), (0,4.5), (-2,0)] #T | |
|
|
2184 | channelPositions = [(4.5,2), (2,4.5), (2,2), (2,0), (0,2)] #Estrella | |
|
|
2185 | ||
|
|
2186 | pairslist0, distances = meteorOps.getPhasePairs(channelPositions) | |
|
|
2187 | h = (hmin,hmax) | |
|
|
2188 | ||
|
|
2189 | arrayParameters = meteorOps.getMeteorParams(arrayParameters, azimuth, h, pairsList, distances, jph) | |
|
|
2283 | 2190 | |
|
|
2284 | dataOut.utctimeInit = dataOut.utctime | |
|
|
2285 | dataOut.outputInterval = dataOut.timeInterval | |
|
|
2191 | dataOut.data_param = arrayParameters | |
|
|
2286 | 2192 | return |
|
|
2287 | ||
|
|
2193 | ||
|
|
2288 | 2194 | class PhaseCalibration(Operation): |
|
|
2289 | 2195 | |
|
|
2290 | 2196 | __buffer = None |
| @@ -2709,4 +2615,108 class MeteorOperations(): | |||
|
|
2709 | 2615 | # pairslist = [(channelCentX, channel25X),(channelCentX, channel20X),(channelCentY,channel25Y),(channelCentY, channel20Y)] |
|
|
2710 | 2616 | pairslist = [(chanCX, chan1X),(chanCX, chan2X),(chanCY,chan1Y),(chanCY, chan2Y)] |
|
|
2711 | 2617 | |
|
|
2712 | return pairslist, distances No newline at end of file | |
|
|
2618 | return pairslist, distances | |
|
|
2619 | # def __getAOA(self, phases, pairsList, error, AOAthresh, azimuth): | |
|
|
2620 | # | |
|
|
2621 | # arrayAOA = numpy.zeros((phases.shape[0],3)) | |
|
|
2622 | # cosdir0, cosdir = self.__getDirectionCosines(phases, pairsList) | |
|
|
2623 | # | |
|
|
2624 | # arrayAOA[:,:2] = self.__calculateAOA(cosdir, azimuth) | |
|
|
2625 | # cosDirError = numpy.sum(numpy.abs(cosdir0 - cosdir), axis = 1) | |
|
|
2626 | # arrayAOA[:,2] = cosDirError | |
|
|
2627 | # | |
|
|
2628 | # azimuthAngle = arrayAOA[:,0] | |
|
|
2629 | # zenithAngle = arrayAOA[:,1] | |
|
|
2630 | # | |
|
|
2631 | # #Setting Error | |
|
|
2632 | # #Number 3: AOA not fesible | |
|
|
2633 | # indInvalid = numpy.where(numpy.logical_and((numpy.logical_or(numpy.isnan(zenithAngle), numpy.isnan(azimuthAngle))),error == 0))[0] | |
|
|
2634 | # error[indInvalid] = 3 | |
|
|
2635 | # #Number 4: Large difference in AOAs obtained from different antenna baselines | |
|
|
2636 | # indInvalid = numpy.where(numpy.logical_and(cosDirError > AOAthresh,error == 0))[0] | |
|
|
2637 | # error[indInvalid] = 4 | |
|
|
2638 | # return arrayAOA, error | |
|
|
2639 | # | |
|
|
2640 | # def __getDirectionCosines(self, arrayPhase, pairsList): | |
|
|
2641 | # | |
|
|
2642 | # #Initializing some variables | |
|
|
2643 | # ang_aux = numpy.array([-8,-7,-6,-5,-4,-3,-2,-1,0,1,2,3,4,5,6,7,8])*2*numpy.pi | |
|
|
2644 | # ang_aux = ang_aux.reshape(1,ang_aux.size) | |
|
|
2645 | # | |
|
|
2646 | # cosdir = numpy.zeros((arrayPhase.shape[0],2)) | |
|
|
2647 | # cosdir0 = numpy.zeros((arrayPhase.shape[0],2)) | |
|
|
2648 | # | |
|
|
2649 | # | |
|
|
2650 | # for i in range(2): | |
|
|
2651 | # #First Estimation | |
|
|
2652 | # phi0_aux = arrayPhase[:,pairsList[i][0]] + arrayPhase[:,pairsList[i][1]] | |
|
|
2653 | # #Dealias | |
|
|
2654 | # indcsi = numpy.where(phi0_aux > numpy.pi) | |
|
|
2655 | # phi0_aux[indcsi] -= 2*numpy.pi | |
|
|
2656 | # indcsi = numpy.where(phi0_aux < -numpy.pi) | |
|
|
2657 | # phi0_aux[indcsi] += 2*numpy.pi | |
|
|
2658 | # #Direction Cosine 0 | |
|
|
2659 | # cosdir0[:,i] = -(phi0_aux)/(2*numpy.pi*0.5) | |
|
|
2660 | # | |
|
|
2661 | # #Most-Accurate Second Estimation | |
|
|
2662 | # phi1_aux = arrayPhase[:,pairsList[i][0]] - arrayPhase[:,pairsList[i][1]] | |
|
|
2663 | # phi1_aux = phi1_aux.reshape(phi1_aux.size,1) | |
|
|
2664 | # #Direction Cosine 1 | |
|
|
2665 | # cosdir1 = -(phi1_aux + ang_aux)/(2*numpy.pi*4.5) | |
|
|
2666 | # | |
|
|
2667 | # #Searching the correct Direction Cosine | |
|
|
2668 | # cosdir0_aux = cosdir0[:,i] | |
|
|
2669 | # cosdir0_aux = cosdir0_aux.reshape(cosdir0_aux.size,1) | |
|
|
2670 | # #Minimum Distance | |
|
|
2671 | # cosDiff = (cosdir1 - cosdir0_aux)**2 | |
|
|
2672 | # indcos = cosDiff.argmin(axis = 1) | |
|
|
2673 | # #Saving Value obtained | |
|
|
2674 | # cosdir[:,i] = cosdir1[numpy.arange(len(indcos)),indcos] | |
|
|
2675 | # | |
|
|
2676 | # return cosdir0, cosdir | |
|
|
2677 | # | |
|
|
2678 | # def __calculateAOA(self, cosdir, azimuth): | |
|
|
2679 | # cosdirX = cosdir[:,0] | |
|
|
2680 | # cosdirY = cosdir[:,1] | |
|
|
2681 | # | |
|
|
2682 | # zenithAngle = numpy.arccos(numpy.sqrt(1 - cosdirX**2 - cosdirY**2))*180/numpy.pi | |
|
|
2683 | # azimuthAngle = numpy.arctan2(cosdirX,cosdirY)*180/numpy.pi + azimuth #0 deg north, 90 deg east | |
|
|
2684 | # angles = numpy.vstack((azimuthAngle, zenithAngle)).transpose() | |
|
|
2685 | # | |
|
|
2686 | # return angles | |
|
|
2687 | # | |
|
|
2688 | # def __getHeights(self, Ranges, zenith, error, minHeight, maxHeight): | |
|
|
2689 | # | |
|
|
2690 | # Ramb = 375 #Ramb = c/(2*PRF) | |
|
|
2691 | # Re = 6371 #Earth Radius | |
|
|
2692 | # heights = numpy.zeros(Ranges.shape) | |
|
|
2693 | # | |
|
|
2694 | # R_aux = numpy.array([0,1,2])*Ramb | |
|
|
2695 | # R_aux = R_aux.reshape(1,R_aux.size) | |
|
|
2696 | # | |
|
|
2697 | # Ranges = Ranges.reshape(Ranges.size,1) | |
|
|
2698 | # | |
|
|
2699 | # Ri = Ranges + R_aux | |
|
|
2700 | # hi = numpy.sqrt(Re**2 + Ri**2 + (2*Re*numpy.cos(zenith*numpy.pi/180)*Ri.transpose()).transpose()) - Re | |
|
|
2701 | # | |
|
|
2702 | # #Check if there is a height between 70 and 110 km | |
|
|
2703 | # h_bool = numpy.sum(numpy.logical_and(hi > minHeight, hi < maxHeight), axis = 1) | |
|
|
2704 | # ind_h = numpy.where(h_bool == 1)[0] | |
|
|
2705 | # | |
|
|
2706 | # hCorr = hi[ind_h, :] | |
|
|
2707 | # ind_hCorr = numpy.where(numpy.logical_and(hi > minHeight, hi < maxHeight)) | |
|
|
2708 | # | |
|
|
2709 | # hCorr = hi[ind_hCorr] | |
|
|
2710 | # heights[ind_h] = hCorr | |
|
|
2711 | # | |
|
|
2712 | # #Setting Error | |
|
|
2713 | # #Number 13: Height unresolvable echo: not valid height within 70 to 110 km | |
|
|
2714 | # #Number 14: Height ambiguous echo: more than one possible height within 70 to 110 km | |
|
|
2715 | # | |
|
|
2716 | # indInvalid2 = numpy.where(numpy.logical_and(h_bool > 1, error == 0))[0] | |
|
|
2717 | # error[indInvalid2] = 14 | |
|
|
2718 | # indInvalid1 = numpy.where(numpy.logical_and(h_bool == 0, error == 0))[0] | |
|
|
2719 | # error[indInvalid1] = 13 | |
|
|
2720 | # | |
|
|
2721 | # return heights, error | |
|
|
2722 | No newline at end of file | |
General Comments 0
You need to be logged in to leave comments.
Login now