-
Notifications
You must be signed in to change notification settings - Fork 14
/
Copy pathsrsChEstimatorUnittest.m
1032 lines (905 loc) · 48 KB
/
srsChEstimatorUnittest.m
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
%srsChEstimatorUnittest Unit tests for the port channel estimator.
% This class implements unit tests for the port channel estimator functions using
% the matlab.unittest framework. The simplest use consists in creating an object with
% testCase = srsChEstimatorUnittest
% and then running all the tests with
% testResults = testCase.run
%
% srsChEstimatorUnittest Properties (Constant):
%
% srsBlock - The tested block (i.e., 'port_channel_estimator').
% srsBlockType - The type of the tested block, including layer
% (i.e., 'phy/upper/signal_processors').
%
% srsChEstimatorUnittest Properties (ClassSetupParameter):
%
% outputPath - Path to the folder where the test results are stored.
%
% srsChEstimatorUnittest Properties (TestParameter):
%
% configuration - Description of the allocated REs and DM-RS pattern.
% SubcarrierSpacing - Subcarrier spacing in kHz.
% NumLayers - Number of transmission layers.
% FrequencyHopping - Frequency hopping type.
% CarrierOffset - Carrier frequency offset, as a fraction of the subcarrier spacing.
%
% srsChEstimatorUnittest Methods:
%
% characterize - Draws the empircical MSE performance curve of the estimator.
%
% srsChEstimatorUnittest Methods (TestTags = {'testvector'}):
%
% testvectorGenerationCases - Generates a test vector according to the provided
% parameters.
%
% srsChEstimatorUnittest Methods (TestTags = {'testmex'}):
%
% compareMex - Compares mex results with those from the reference estimator.
%
% srsChEstimatorUnittest Methods (Access = protected):
%
% addTestIncludesToHeaderFile - Adds include directives to the test header file.
% addTestDefinitionToHeaderFile - Adds details (e.g., type/variable declarations)
% to the test header file.
%
% See also matlab.unittest.
% Copyright 2021-2025 Software Radio Systems Limited
%
% This file is part of srsRAN-matlab.
%
% srsRAN-matlab is free software: you can redistribute it and/or
% modify it under the terms of the BSD 2-Clause License.
%
% srsRAN-matlab is distributed in the hope that it will be useful,
% but WITHOUT ANY WARRANTY; without even the implied warranty of
% MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
% BSD 2-Clause License for more details.
%
% A copy of the BSD 2-Clause License can be found in the LICENSE
% file in the top-level directory of this distribution.
classdef srsChEstimatorUnittest < srsTest.srsBlockUnittest
properties (Constant)
%Name of the tested block.
srsBlock = 'port_channel_estimator'
%Type of the tested block, including layers.
srsBlockType = 'phy/upper/signal_processors'
end % of properties (Constant)
properties (Hidden, Constant)
% Number of resource elements in a RB and OFDM symbols in a slot.
NRE = 12
nSymbolsSlot = 14
end % of properties (Hidden, Constant)
properties (Hidden, SetAccess=private)
% Fix BWP size and start as well as the frame number, since they
% are irrelevant for the test.
NSizeBWP = 51
NStartBWP = 1
NSizeGrid = 52 % srsChEstimatorUnittest.NSizeBWP + srsChEstimatorUnittest.NStartBWP
end % of properties (Hidden, SetAccess=private)
properties (ClassSetupParameter)
%Path to results folder (old 'port_channel_estimator' tests will be erased).
outputPath = {['testChEstimator', char(datetime('now', 'Format', 'yyyyMMdd''T''HHmmss'))]}
end
properties (TestParameter)
%Configuration.
% A configuration structure array with fields:
% nPRBs - Number of allocated PRBs (0...51)
% symbolAllocation - A two-element array denoting the first allocated OFDM symbol (0...13)
% and the number of allocated OFDM symbols (1...14).
% dmrsOffset - Number of non-DM-RS REs at the beginning of the RB (0, 1).
% dmrsStrideSCS - DM-RS frequency domain stride (1, 2, 3), that is the distance
% between two consecutive DM-RS REs (distance of 1 being back-to-back).
% dmrsStrideTime - Stride between OFDM symbols containing DM-RS (1, 2, 4, 6, 20).
% Use 20 for a single DM-RS symbol.
% betaDMRS - The gain of the DM-RS pilots with respect to the data
% symbols in dB (0, 3).
% supportMultiLayer - Boolean flag that specifies whether the configuration supports
% multiple transmission layers (true) or not (false).
% smoothing - The frequency-domain smoothing strategy to be used with the
% current configuration ('filter', 'mean', 'none').
% cfocompensate - A boolean flag denoting whether the channel estimator should
% compensate (true) or not (false) the CFO.
configuration = {...
struct(... % #1: PUSCH DM-RS configuration Type 1 (inspired to).
'nPRBs', 3, ...
'symbolAllocation', [0, 14], ...
'dmrsOffset', 0, ...
'dmrsStrideSCS', 2, ...
'dmrsStrideTime', 4, ...
'betaDMRS', -3, ...
'supportMultiLayer', true, ...
'smoothing', 'filter', ...
'cfocompensate', true ...
),...
struct(... % #2: PUSCH DM-RS configuration Type 1 (inspired to).
'nPRBs', 20, ...
'symbolAllocation', [0, 14], ...
'dmrsOffset', 0, ...
'dmrsStrideSCS', 2, ...
'dmrsStrideTime', 4, ...
'betaDMRS', -3, ...
'supportMultiLayer', true, ...
'smoothing', 'filter', ...
'cfocompensate', true ...
), ...
struct(... % #3: PUSCH DM-RS configuration Type 1 (inspired to).
'nPRBs', 51, ...
'symbolAllocation', [0, 14], ...
'dmrsOffset', 0, ...
'dmrsStrideSCS', 2, ...
'dmrsStrideTime', 4, ...
'betaDMRS', -3, ...
'supportMultiLayer', true, ...
'smoothing', 'filter', ...
'cfocompensate', true ...
), ...
struct(... % #4: PUCCH Format 1 (inspired to).
'nPRBs', 1, ...
'symbolAllocation', [8, 4], ...
'dmrsOffset', 0, ...
'dmrsStrideSCS', 1, ...
'dmrsStrideTime', 2, ...
'betaDMRS', 0, ...
'supportMultiLayer', false, ...
'smoothing', 'mean', ...
'cfocompensate', false ...
), ...
struct(... % #5: PUCCH Format 1 (inspired to).
'nPRBs', 1, ...
'symbolAllocation', [0, 14], ...
'dmrsOffset', 0, ...
'dmrsStrideSCS', 1, ...
'dmrsStrideTime', 2, ...
'betaDMRS', 0, ...
'supportMultiLayer', false, ...
'smoothing', 'mean', ...
'cfocompensate', false ...
), ...
struct(... % #6: PUCCH Format 2 (inspired to).
'nPRBs', 1, ...
'symbolAllocation', [0, 2], ...
'dmrsOffset', 1, ...
'dmrsStrideSCS', 3, ...
'dmrsStrideTime', 1, ...
'betaDMRS', 0, ...
'supportMultiLayer', false, ...
'smoothing', 'filter', ...
'cfocompensate', true ...
), ...
struct(... % #7: PUCCH Format 2 (inspired to).
'nPRBs', 6, ...
'symbolAllocation', [5, 1], ...
'dmrsOffset', 1, ...
'dmrsStrideSCS', 3, ...
'dmrsStrideTime', 1, ...
'betaDMRS', 0, ...
'supportMultiLayer', false, ...
'smoothing', 'filter', ...
'cfocompensate', true ...
), ...
struct(... % #8: PUCCH Format 2 (inspired to).
'nPRBs', 16, ...
'symbolAllocation', [5, 2], ...
'dmrsOffset', 1, ...
'dmrsStrideSCS', 3, ...
'dmrsStrideTime', 1, ...
'betaDMRS', 0, ...
'supportMultiLayer', false, ...
'smoothing', 'filter', ...
'cfocompensate', true ...
), ...
}
%Subcarrier spacing in kHz.
SubcarrierSpacing = {15, 30}
%Number of transmission layers.
NumLayers = {1, 2, 3, 4}
%Frequency hopping type ('neither', 'intraSlot').
% Note: Interslot frequency hopping is currently not considered.
FrequencyHopping = {'neither', 'intraSlot'}
%Carrier frequency offset, as a fraction of the subcarrier spacing.
CarrierOffset = {0, 0.007, -0.013, 0.027}
end % of properties (TestParameter)
properties (Hidden)
%OFDM symbol in which the second hop starts (if any).
secondHop
%Mask of OFDM symbols carrying DM-RS.
DMRSsymbols
%Mask of REs carrying DM-RS (relative to one PRB and one OFDM symbol).
DMRSREmask
end % of properties (Hidden)
methods (Access = protected)
function addTestIncludesToHeaderFile(~, fileID)
%addTestIncludesToHeaderFile(OBJ, FILEID) adds include directives to
% the header file pointed by FILEID, which describes the test vectors.
fprintf(fileID, '#include "../../support/resource_grid_test_doubles.h"\n');
fprintf(fileID, '#include "srsran/phy/upper/signal_processors/port_channel_estimator.h"\n');
fprintf(fileID, '#include "srsran/phy/upper/signal_processors/port_channel_estimator_parameters.h"\n');
fprintf(fileID, '#include "srsran/support/file_vector.h"\n');
fprintf(fileID, '#include <optional>\n');
end
function addTestDefinitionToHeaderFile(~, fileID)
%addTestDefinitionToHeaderFile(OBJ, FILEID) adds test details (e.g., type
% and variable declarations) to the header file pointed by FILEID, which
% describes the test vectors.
fprintf(fileID, 'struct test_case_t {\n');
fprintf(fileID, ' port_channel_estimator::configuration cfg;\n');
fprintf(fileID, ' port_channel_estimator_fd_smoothing_strategy smoothing;\n');
fprintf(fileID, ' bool compensate_cfo = false;\n');
fprintf(fileID, ' unsigned grid_size_prbs = 0;\n');
fprintf(fileID, ' float rsrp = 0;\n');
fprintf(fileID, ' float epre = 0;\n');
fprintf(fileID, ' float snr_true = 0;\n');
fprintf(fileID, ' float snr_est = 0;\n');
fprintf(fileID, ' float noise_var_est = 0;\n');
fprintf(fileID, ' float ta_us = 0;\n');
fprintf(fileID, ' float cfo_true_Hz = 0;\n');
fprintf(fileID, ' std::optional<float> cfo_est_Hz = 0;\n');
fprintf(fileID, ' file_vector<resource_grid_reader_spy::expected_entry_t> grid;\n');
fprintf(fileID, ' file_vector<cf_t> pilots;\n');
fprintf(fileID, ' file_vector<resource_grid_reader_spy::expected_entry_t> estimates;\n');
fprintf(fileID, '};\n');
end
end % of methods (Access = protected)
methods (Test, TestTags = {'testvector'})
function testvectorGenerationCases(obj, configuration, SubcarrierSpacing, ...
NumLayers, FrequencyHopping, CarrierOffset)
%testvectorGenerationCases - Generates a test vector according to the provided
% CONFIGURATION, SUBCARRIERSPACING, NUMLAYERS, FREQUENCYHOPPING type and CARRIEROFFSET.
import srsTest.helpers.writeResourceGridEntryFile
import srsTest.helpers.writeComplexFloatFile
[fullConfig, channel, receivedRG, results] = obj.configureAndMatlab(configuration, ...
SubcarrierSpacing, NumLayers, FrequencyHopping, CarrierOffset);
% Generate a unique test ID.
testID = obj.generateTestID;
channelEst = results.ChannelEst;
noiseEst = results.NoiseEst;
rsrp = results.RSRP;
epre = results.EPRE;
timeAlignment = results.TimeAlignment;
cfoEst = results.CFO;
% TODO: The ratio of the two quantities below should give a metric that allows us
% to decide whether pilots were sent or not. However, it should be normalized
% and it's a bit tricky.
% detectMetricNum = detectMetricNum / nDMRSsymbols;
% detectMetricDen = noiseEst;
% detectionMetric = detectMetricNum / detectMetricDen;
SNR = channel.SNR;
channelRG = channel.RG;
noiseVar = 10^(-SNR/10);
channelDelay = channel.Delay;
cfo = channel.CFO;
betaDMRS = fullConfig.BetaDMRS;
fftSize = fullConfig.FFTSize;
snrEst = rsrp * NumLayers / betaDMRS^2 / noiseEst;
% A few very loose checks, just to ensure we are not completely out of place.
if (configuration.nPRBs > 2)
chEstIdx = (channelEst ~= 0);
obj.assertEqual(channelEst(chEstIdx), channelRG(chEstIdx), "Wrong channel coefficients.", RelTol = 0.4);
obj.assertEqual(noiseEst, noiseVar, "Wrong noise variance.", RelTol = 1.2 * NumLayers);
obj.assertEqual(snrEst, 10^(SNR/10), "Wrong SNR.", RelTol = 2.5);
obj.assertEqual(timeAlignment, channelDelay / fftSize / SubcarrierSpacing / 1000, ...
"Wrong time alignment.", AbsTol = 2e-7);
cfoHz = cfo * SubcarrierSpacing * 1000;
if (~isempty(cfoEst) && (abs(cfoEst - cfoHz) > 40) && ((cfoHz == 0) || abs(cfoEst / cfoHz - 1) > 0.7))
warning('srsran_matlab:srsChEstimatorUnittest', 'Estimated CFO = %f, True CFO = %f.', cfoEst, cfoHz);
end
end
% Write the received resource grid.
[scs, syms, vals] = find(receivedRG);
obj.saveDataFile('_test_input_rg', testID, @writeResourceGridEntryFile, ...
vals, [scs, syms, zeros(length(scs), 1)] - 1);
% Write the estimated channel.
[scs, syms, vals] = find(channelEst);
obj.saveDataFile('_test_output_ch_est', testID, @writeResourceGridEntryFile, ...
vals, [scs - 1, mod(syms - 1, 14), floor((syms - 1) / 14)]);
pilots = fullConfig.Pilots;
% Write the pilots.
obj.saveDataFile('_test_pilots', testID, @writeComplexFloatFile, pilots(:));
hop1 = fullConfig.Hop1;
hop2 = fullConfig.Hop2;
dmrsPattern = {...
obj.DMRSsymbols, ... % symbols
hop1.maskPRBs, ... % rb_mask
hop2.maskPRBs, ... % rb_mask2
obj.secondHop, ... % hopping_symbol_index
obj.DMRSREmask, ... % re_pattern
};
if (NumLayers >= 2)
% Layer 1 shares the same resources as layer 0.
dmrsPattern = {dmrsPattern dmrsPattern};
end
if (NumLayers >= 3)
% Layer 2 shares the same time resources as layers 0 and 1,
% but the RE mask is negated.
dmrsPattern = [dmrsPattern dmrsPattern(end)];
dmrsPattern{end}{end} = ~dmrsPattern{end}{end};
end
if (NumLayers == 4)
% Layer 3 shares the same resources as layer 2.
dmrsPattern = [dmrsPattern dmrsPattern(end)];
end
startSymbol = configuration.symbolAllocation(1);
nAllocatedSymbols = configuration.symbolAllocation(2);
scsString = sprintf('subcarrier_spacing::kHz%d', SubcarrierSpacing);
configurationOut = {...
scsString, ... % scs
'cyclic_prefix::NORMAL', ... % cp
startSymbol, ... % first_symbol
nAllocatedSymbols, ... % nof_symbols
{dmrsPattern}, ... % dmrs_patterns
{0}, ... % rx_ports
betaDMRS, ... % betaDMRS
};
if isempty(cfoEst)
cfoEst = {};
end
smoothingOut = ['port_channel_estimator_fd_smoothing_strategy::' configuration.smoothing];
context = {...
configurationOut, ...
smoothingOut, ...
configuration.cfocompensate, ...
obj.NSizeGrid, ...
rsrp, ...
epre, ...
SNR, ...
10 * log10(snrEst), ...
noiseEst, ...
timeAlignment * 1e6, ...
cfo * SubcarrierSpacing * 1000, ...
cfoEst, ...
};
testCaseString = obj.testCaseToString(testID, context, false, ...
'_test_input_rg', '_test_pilots', '_test_output_ch_est');
% Add the test to the header file.
obj.addTestToHeaderFile(obj.headerFileID, testCaseString);
end % of function testvectorGenerationCases(...)
end % of methods (Test, TestTags = {'testvector'})
methods (Test, TestTags = {'testmex'})
function compareMex(obj, configuration, SubcarrierSpacing, NumLayers, FrequencyHopping, CarrierOffset)
%compareMex - Compare mex results with those from the reference estimator for
% a given CONFIGURATION, SUBCARRIERSPACING, FREQUENCYHOPPING type and CARRIEROFFSET.
import srsMEX.phy.srsMultiPortChannelEstimator
capabilitiesMEX = srsMEX.phy.srsPUSCHCapabilitiesMEX();
obj.assumeGreaterThanOrEqual(capabilitiesMEX.NumLayers, NumLayers, ...
sprintf('The current MEX version only works with max. %d layers, requested %d.', capabilitiesMEX.NumLayers, NumLayers));
[fullConfig, ~, receivedRG, results] = configureAndMatlab(obj, ...
configuration, SubcarrierSpacing, NumLayers, FrequencyHopping, CarrierOffset);
hop1 = fullConfig.Hop1;
hop2 = fullConfig.Hop2;
pilots = fullConfig.Pilots;
betaDMRS = fullConfig.BetaDMRS;
channelEst = results.ChannelEst;
noiseEst = results.NoiseEst;
rsrp = results.RSRP;
epre = results.EPRE;
timeAlignment = results.TimeAlignment;
cfoEst = results.CFO;
% Cast input for the mex estimator.
pilotRBMask = hop1.maskPRBs * hop1.DMRSsymbols';
if (~isempty(hop2.maskPRBs) && ~isempty(hop2.DMRSsymbols))
pilotRBMask = pilotRBMask + hop2.maskPRBs * hop2.DMRSsymbols';
end
nPRB = size(hop1.maskPRBs, 1);
nSymbols = size(hop1.DMRSsymbols, 1);
[nRE, nCDM] = size(hop1.DMRSREmask);
pilotMask = nan(nPRB * nRE, nSymbols, nCDM);
nPilots = sum(pilotRBMask, 'all') * sum(hop1.DMRSREmask(:, 1));
pilotIndices = nan(nPilots, NumLayers);
for iLayer = 1:2:NumLayers
iCDM = floor(iLayer / 2) + 1;
pilotMask(:, :, iCDM) = kron(pilotRBMask, hop1.DMRSREmask(:, iCDM));
pilotIndices(:, iLayer) = find(pilotMask(:, :, iCDM)) + prod(size(receivedRG, [1, 2])) * (iLayer - 1);
if (iLayer + 1 <= NumLayers)
pilotIndices(:, iLayer + 1) = pilotIndices(:, iLayer) + prod(size(receivedRG, [1, 2]));
end
end
mexEstimator = srsMultiPortChannelEstimator(...
Smoothing=configuration.smoothing, ...
CompensateCFO=configuration.cfocompensate ...
);
[channelEstMEX, noiseEstMEX, extra] ...
= mexEstimator(receivedRG, configuration.symbolAllocation, pilotIndices, reshape(pilots, [], NumLayers), ...
SubcarrierSpacing=SubcarrierSpacing, ...
HoppingIndex=hop2.startSymbol, ...
BetaScaling=betaDMRS ...
);
% The tolerance for the time alignment is one timing-advance step size
% (32 samples with a 4096-point DFT).
toleranceTA = 32000 / (4096 * SubcarrierSpacing);
chEstIdx = (channelEst ~= 0);
obj.assertEqual(channelEstMEX(chEstIdx), channelEst(chEstIdx), 'Wrong channel estimates.', AbsTol = 0.008);
obj.assertEqual(noiseEstMEX, noiseEst, 'Wrong noise variance estimate.', AbsTol = 5e-4);
obj.assertEqual(extra.RSRP, rsrp, 'Wrong RSRP estimate.', AbsTol = 5e-4);
obj.assertEqual(extra.EPRE, epre, 'Wrong EPRE estimate.', AbsTol = 5e-4);
obj.assertEqual(extra.SINR, rsrp * NumLayers / betaDMRS^2 / noiseEst, 'Wrong SINR estimate.', RelTol = 0.004);
obj.assertEqual(extra.TimeAlignment, timeAlignment, 'Wrong time alignment estimate.', AbsTol = toleranceTA);
obj.assertEqual(extra.CFO, cfoEst, 'Wrong CFO.', RelTol = 0.04);
end % of function compareMex(...)
end % of methods (Test, TestTags = {'testmex'})
methods % public
function [mse, noiseEst, rsrpEst, epreEst, cfoEst, crlb] = characterize(obj, configuration, ...
FrequencyHopping, scs, nLayers, channelType, delay, doppler, cfo, snrValues, nRuns, sizeBWP)
%characterize - Draw the empirical MSE performance curve of the estimator.
% MSE = characterize(OBJ, CONFIGURATION, FREQUENCYHOPPING, SCS, NLAYERS, CHANNELTYPE, DELAY, DOPPLER, CFO, SNRVALUES, NRUNS)
% returns the empirical mean squared error of the channel estimation after NRUNS simulations
% and for all SNRVALUES. CONFIGURATION, FREQUENCYHOPPING, SCS and NLAYERS provide the physical
% channel configuration and CHANNELTYPE, DELAY, DOPPLER and CFO specify the simulated channel model.
% Note: DELAY is the delay spread for the IEEE channel types, and the Path delay
% for the 'pure-delay' channel type.
% Note: DOPPLER is the maximum Doppler shift in hertz (effect of scattering), not to
% be confused with the CFO.
%
% MSE = characterize(..., BWP) also changes the BWP size (expressed as a number of RBs).
% The default BWP size is 51.
%
% [MSE, NOISEEST, RSRPEST, EPREEST, CFOEST, CRLB] = characterize(...) also returns the
% estimates of noise variance, RSRP, EPRE and CFO for all runs and all SNR values,
% as well as the CRLB for the channel estimation. The CRLB is computed assuming
% the entire band is available for estimation, with pilots positioned with
% the same pattern as the DM-RS (first column) or with pilots in all REs
% (second column).
%
% For CONFIGURATION and FREQUENCYHOPPING, see <a href="matlab:help srsChEstimatorUnittest">the main class documentation</a>.
% SNRVALUES is an array of SNR values in decibel.
% NRUNS is an integer number of simulations.
arguments
obj (1, 1) srsChEstimatorUnittest
configuration (1, 1) struct {mustBeConfiguration}
FrequencyHopping (1, :) char {mustBeMember(FrequencyHopping, {'neither', 'intraSlot'})}
scs (1, 1) double {mustBeMember(scs, [15, 30])}
nLayers (1, 1) double {mustBeMember(nLayers, 1:4)}
channelType (1, :) char {mustBeMember(channelType, {'pure-delay', ...
'TDL-A', 'TDL-B', 'TDL-C', 'TDL-D', 'TDL-E', ...
'TDLA30', 'TDLB100', 'TDLC300', 'TDLC60'})}
delay (1, 1) double {mustBeReal, mustBeNonnegative}
doppler (1, 1) double {mustBeReal, mustBeNonnegative}
cfo (1, 1) double {mustBeReal}
snrValues double {mustBeReal, mustBeVector}
nRuns (1, 1) double {mustBeNonnegative, mustBeInteger}
sizeBWP (1, 1) double = NaN
end
import srsLib.phy.upper.signal_processors.srsChannelEstimator
if ~isempty(delay)
validateattributes(delay, 'double', {'nonnegative'});
end
if ~isnan(sizeBWP)
validateattributes(sizeBWP, 'double', {'positive', 'integer', '<=', 273});
obj.NSizeBWP = sizeBWP;
obj.NSizeGrid = obj.NStartBWP + obj.NSizeBWP;
end
fullConfig = configureAndMatlab(obj, configuration, scs, nLayers, FrequencyHopping, []);
hop1 = fullConfig.Hop1;
hop2 = fullConfig.Hop2;
pilots = fullConfig.Pilots;
betaDMRS = fullConfig.BetaDMRS;
% Configure carrier.
carrier = nrCarrierConfig;
carrier.CyclicPrefix = 'Normal';
carrier.SubcarrierSpacing = scs; % kHz
carrier.NSlot = 0;
carrier.NSizeGrid = obj.NSizeGrid;
waveformInfo = nrOFDMInfo(carrier);
channel = configureChannel(channelType, delay, doppler, waveformInfo.SampleRate, ...
carrier.SubcarrierSpacing, nLayers);
% Place pilots on the resource grid.
transmittedRG = obj.transmitPilots(pilots, betaDMRS, hop1, hop2);
transmittedWF = nrOFDMModulate(carrier, transmittedRG);
nEstSCS = hop1.nPRBs * obj.NRE;
if ~isempty(hop2.maskPRBs)
nEstSCS = nEstSCS * 2;
end
mse = zeros(length(snrValues), nEstSCS, nLayers);
noiseEst = zeros(length(snrValues), nRuns);
rsrpEst = zeros(length(snrValues), nRuns);
epreEst = zeros(length(snrValues), nRuns);
cfoEst = zeros(length(snrValues), nRuns);
CPDurations = waveformInfo.CyclicPrefixLengths(1:14);
CPDurations = CPDurations / sum(CPDurations) / scs;
% Configure estimator.
EstimatorConfig.nPilotsNoiseAvg = sum(obj.DMRSREmask);
EstimatorConfig.scs = scs * 1000; % SCS in hertz
EstimatorConfig.CyclicPrefixDurations = CPDurations;
EstimatorConfig.Smoothing = configuration.smoothing;
EstimatorConfig.CFOCompensate = configuration.cfocompensate;
for iRun = 1:nRuns
reset(channel);
[receivedWF0, pathGains, sampleTimes] = channel(transmittedWF);
if cfo ~= 0
nSamples = size(receivedWF0, 1);
if (~exist('cfoPhase', 'var') || length(cfoPhase) ~= nSamples)
timeIx = (0:length(receivedWF0)-1).';
cfoPhase = exp(2j * pi * timeIx * cfo / waveformInfo.SampleRate);
end
receivedWF0 = receivedWF0 .* cfoPhase;
end
noise0 = randn(size(receivedWF0)) + 1j * randn(size(receivedWF0));
iSNR = 0;
for SNR = snrValues
iSNR = iSNR + 1;
noiseVar = 10^(-SNR/10) / waveformInfo.Nfft;
noise = noise0 * sqrt(noiseVar / 2);
receivedWF = receivedWF0 + noise;
% Compute received resource grid.
receivedRG = nrOFDMDemodulate(carrier, receivedWF);
[channelEst, noiseEstL, rsrpEstL, epreEstL, ~, cfoEstL] ...
= srsChannelEstimator(receivedRG, pilots, betaDMRS, hop1, hop2, EstimatorConfig);
noiseEst(iSNR, iRun) = noiseEstL;
rsrpEst(iSNR, iRun) = rsrpEstL;
epreEst(iSNR, iRun) = epreEstL;
cfoEst(iSNR, iRun) = cfoEstL;
% Get the true channel, for comparison.
pathFilters = channel.getPathFilters();
channelTrue = squeeze(nrPerfectChannelEstimate(carrier, pathGains, pathFilters, 0, sampleTimes));
if cfo ~= 0
cfoNorm = cfo / scs / 1000;
cfoFreq = [waveformInfo.CyclicPrefixLengths(1) waveformInfo.CyclicPrefixLengths(2:14) + waveformInfo.Nfft];
cfoFreq = cumsum(cfoFreq) * cfoNorm / waveformInfo.Nfft;
cfoFreq = exp(2j * pi * cfoFreq);
for iLayer = 1:nLayers
channelTrue(:, :, iLayer) = channelTrue(:, :, iLayer) * diag(cfoFreq) * exp(1j * pi * ((waveformInfo.Nfft - 1) / waveformInfo.Nfft) * cfoNorm);
end
end
% Just for debugging/analysis purposes: set to true to visualize
% the effect of channel and channel estimation on a random QAM
% points.
if false
iLayer = 1; %#ok<UNRCH>
whatSymbol = 1+hop1.startSymbol;
whatSCS = (channelEst(:, whatSymbol, iLayer) ~= 0);
nSCS = sum(whatSCS);
% Create some random QAM points.
fakeSymbols = srsTest.helpers.randmod('QPSK', [nSCS, 50]);
% Apply the true channel and ZF-equalize with the estimated channel (SC-wise).
rr = diag(channelTrue(whatSCS, whatSymbol, iLayer) ./ channelEst(whatSCS, whatSymbol, iLayer)) * fakeSymbols;
if (size(rr, 1) >= 60)
% Split edge and middle points, to visualize the difference in
% the estimation performance.
rrEdge = rr([1:24, end-23:end], :);
rrMiddle = rr(25:end-24, :);
plot(real(rrEdge(:)), imag(rrEdge(:)), 'rx', real(rrMiddle(:)), imag(rrMiddle(:)), 'bx')
else
plot(real(rr(:)), imag(rr(:)), 'bx')
end
pause
end
scsIdx = obj.NRE * hop1.PRBstart + (1: obj.NRE * hop1.nPRBs);
estErrors = channelEst(scsIdx, hop1.CHsymbols, :) - channelTrue(scsIdx, hop1.CHsymbols, :);
SQestErrors = squeeze(sum(abs(estErrors).^2, 2));
if ~isempty(hop2.maskPRBs)
scsIdx = obj.NRE * hop2.PRBstart + (1: obj.NRE * hop2.nPRBs);
estErrors2 = channelEst(scsIdx, hop2.CHsymbols, :) - channelTrue(scsIdx, hop2.CHsymbols, :);
SQestErrors2 = squeeze(sum(abs(estErrors2).^2, 2));
else
SQestErrors2 = double.empty(0, nLayers);
end
for iLayer = 1:nLayers
mse(iSNR, :, iLayer) = mse(iSNR, :, iLayer) + [SQestErrors(:, iLayer); SQestErrors2(:, iLayer)]' / nRuns;
end
end
end
crlb = repmat(10.^(-snrValues(:)/10), 1, 2) / betaDMRS^2 / sum(hop1.DMRSsymbols);
crlb = (crlb' .* computeCRLB(hop1.maskPRBs, hop1.DMRSREmask(:, 1)))';
end % of function characterize(...)
end % of methods % public
methods (Access = private)
function [hop1, hop2] = configureHops(obj, configuration, NumLayers, FrequencyHopping)
%Creates a description of the resources allocated in each hop.
startSymbol = configuration.symbolAllocation(1);
nAllocatedSymbols = configuration.symbolAllocation(2);
dmrsStrideTime = configuration.dmrsStrideTime;
% Create a mask of the OFDM symbols carrying DM-RS.
obj.DMRSsymbols = false(14, 1);
obj.DMRSsymbols(startSymbol + (1:dmrsStrideTime:nAllocatedSymbols)) = true;
nPRBs = configuration.nPRBs;
dmrsOffset = configuration.dmrsOffset;
dmrsStrideSCS = configuration.dmrsStrideSCS;
% Create a DM-RS pattern from the offset and stride.
obj.DMRSREmask = false(obj.NRE, 1);
obj.DMRSREmask((dmrsOffset + 1):dmrsStrideSCS:end) = true;
if strcmp(FrequencyHopping, 'intraSlot')
PRBstart = randperm(obj.NSizeBWP - nPRBs + 1, 2) - 1 + obj.NStartBWP;
obj.secondHop = startSymbol + floor(nAllocatedSymbols / 2);
hopMask = [true(obj.secondHop, 1); false(obj.nSymbolsSlot - obj.secondHop, 1)];
hop1.DMRSsymbols = (obj.DMRSsymbols & hopMask);
if NumLayers <= 2
hop1.DMRSREmask = obj.DMRSREmask;
else
hop1.DMRSREmask = [obj.DMRSREmask ~obj.DMRSREmask];
end
hop1.PRBstart = PRBstart(1);
hop1.nPRBs = nPRBs;
hop1.maskPRBs = false(obj.NSizeGrid, 1);
hop1.maskPRBs(hop1.PRBstart + (1:nPRBs)) = true;
hop1.startSymbol = startSymbol;
hop1.nAllocatedSymbols = floor(nAllocatedSymbols / 2);
hop1.CHsymbols = false(obj.nSymbolsSlot, 1);
hop1.CHsymbols(hop1.startSymbol + (1:hop1.nAllocatedSymbols)) = true;
hop2.DMRSsymbols = (obj.DMRSsymbols & (~hopMask));
hop2.DMRSREmask = hop1.DMRSREmask;
hop2.PRBstart = PRBstart(2);
hop2.nPRBs = nPRBs;
hop2.maskPRBs = false(obj.NSizeGrid, 1);
hop2.maskPRBs(hop2.PRBstart + (1:nPRBs)) = true;
hop2.startSymbol = obj.secondHop;
hop2.nAllocatedSymbols = ceil(nAllocatedSymbols / 2);
hop2.CHsymbols = false(obj.nSymbolsSlot, 1);
hop2.CHsymbols(hop2.startSymbol + (1:hop2.nAllocatedSymbols)) = true;
else
PRBstart = randi([0, obj.NSizeBWP - nPRBs]) + obj.NStartBWP;
obj.secondHop = 'std::nullopt';
hop1.DMRSsymbols = obj.DMRSsymbols;
if NumLayers <= 2
hop1.DMRSREmask = obj.DMRSREmask;
else
hop1.DMRSREmask = [obj.DMRSREmask ~obj.DMRSREmask];
end
hop1.PRBstart = PRBstart;
hop1.nPRBs = nPRBs;
hop1.maskPRBs = false(obj.NSizeGrid, 1);
hop1.maskPRBs(hop1.PRBstart + (1:nPRBs)) = true;
hop1.startSymbol = startSymbol;
hop1.nAllocatedSymbols = nAllocatedSymbols;
hop1.CHsymbols = false(obj.nSymbolsSlot, 1);
hop1.CHsymbols(hop1.startSymbol + (1:hop1.nAllocatedSymbols)) = true;
hop2.DMRSsymbols = [];
hop2.maskPRBs = {};
hop2.startSymbol = [];
end
end % of function [hop1 hop2] = configureHops()
function transmittedRG = transmitPilots(obj, pilots, betaDMRS, hop1, hop2)
%Places the pilots on the correct REs and with the correct power on the resource grid.
nLayers = size(pilots, 3);
transmittedRG = complex(zeros(obj.NSizeGrid * obj.NRE, obj.nSymbolsSlot, nLayers));
nPilotSymbolsHop1 = sum(hop1.DMRSsymbols);
processHop(hop1, pilots(:, 1:nPilotSymbolsHop1, :));
if ~isempty(hop2.DMRSsymbols)
processHop(hop2, pilots(:, (nPilotSymbolsHop1 + 1):end, :));
end
% Nested functions
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
function processHop(hop_, pilots_)
%Processes the DM-RS corresponding to a single hop.
% Create a mask for all subcarriers carrying DM-RS.
maskPRBs_ = hop_.maskPRBs;
maskREs_ = (kron(maskPRBs_, obj.DMRSREmask) > 0);
nLayersCDM0_ = min(2, nLayers);
transmittedRG(maskREs_, hop_.DMRSsymbols, 1:nLayersCDM0_) = betaDMRS * pilots_(:, :, 1:nLayersCDM0_);
if nLayers > 2
% DM-RS for layers 3 and 4 are sent on the complementary set of REs.
maskREs_ = (kron(maskPRBs_, ~obj.DMRSREmask) > 0);
transmittedRG(maskREs_, hop_.DMRSsymbols, 3:nLayers) = betaDMRS * pilots_(:, :, 3:nLayers);
end
end % of function processHop(hop_, pilots_)
end % of function transmittedRG = transmitPilots(pilots, hop1, hop2)
function [fullConfig, channel, receivedRG, results] = configureAndMatlab(obj, ...
configuration, SubcarrierSpacing, NumLayers, FrequencyHopping, CarrierOffset)
%Computes secondary configuration parameters and, if the number of output arguments
% is larger than one, runs the MATLAB-based channel estimator.
import srsLib.phy.upper.signal_processors.srsChannelEstimator
import srsLib.ran.utils.scs2cps
import srsTest.helpers.approxbf16
obj.assumeFalse(((configuration.nPRBs == obj.NSizeBWP) || (configuration.symbolAllocation(2) == 1)) ...
&& strcmp(FrequencyHopping, 'intraSlot'), ...
'Cannot do frequency hopping if the entire BWP is allocated or if using a single OFDM symbol.');
obj.assumeFalse(~configuration.supportMultiLayer && (NumLayers > 1), ...
'Multiple transmission layers not supported for this configuration.');
assert((sum(configuration.symbolAllocation) <= obj.nSymbolsSlot), ...
'srsran_matlab:srsChEstimatorUnittest', 'Time allocation exceeds slot length.');
% Configure each hop.
[hop1, hop2] = obj.configureHops(configuration, NumLayers, FrequencyHopping);
% Build DM-RS-like pilots.
nDMRSsymbols = sum(obj.DMRSsymbols);
nPilots = configuration.nPRBs * sum(obj.DMRSREmask) * nDMRSsymbols;
pilots = complex(nan(nPilots * NumLayers, 1), nan(nPilots * NumLayers, 1));
pilots(1:nPilots) = (2 * randi([0 1], nPilots, 2) - 1) * [1; 1j] / sqrt(2);
pilots = reshape(pilots, [], nDMRSsymbols, NumLayers);
if NumLayers >= 2
% We only simulate the case corresponding to DM-RS configuration type 1.
% For CDM group 0, i.e. ports {0, 1}, the DM-RS are sent over the same REs
% and the pilots for port 1 are the same as those of port 0 but for a sign
% change in every other pilot.
pilots(:, :, 2) = pilots(:, :, 1);
pilots(2:2:end, :, 2) = -pilots(2:2:end, :, 2);
end
if NumLayers >= 3
% We only simulate the case corresponding to DM-RS configuration type 1.
% The DM-RS for port 2 are the same as those of port 0, just sent on
% a disjoint set of REs.
pilots(:, :, 3) = pilots(:, :, 1);
end
if NumLayers == 4
% We only simulate the case corresponding to DM-RS configuration type 1.
% The DM-RS for port 3 are the same as those of port 1, just sent on
% a disjoint set of REs.
pilots(:, :, 4) = pilots(:, :, 2);
end
betaDMRS = 10^(-configuration.betaDMRS / 20);
fftSize = obj.NSizeGrid * obj.NRE;
fullConfig = struct(...
'Pilots', pilots, ...
'Hop1', hop1, ...
'Hop2', hop2, ...
'BetaDMRS', betaDMRS, ...
'FFTSize', fftSize ...
);
if nargout > 1
% Place pilots on the resource grid.
transmittedRG = obj.transmitPilots(pilots, betaDMRS, hop1, hop2);
cfo = CarrierOffset; % Fraction of the SCS.
CPDurations = scs2cps(SubcarrierSpacing);
receivedRG = complex(zeros(fftSize, obj.nSymbolsSlot));
channelRG = complex(nan(fftSize, obj.nSymbolsSlot, NumLayers));
% For now, consider a single-tap channel (max delay is 1/4 of
% the cyclic prefix length).
channelDelay = randi([0, floor(fftSize * 0.07 * 0.25)]);
for iLayer = 1:NumLayers
channelCoef = exp(2j * pi * rand) / sqrt(NumLayers);
channelTF = fft([zeros(channelDelay, 1); channelCoef; zeros(5, 1)], fftSize);
channelTF = fftshift(channelTF);
% We assume the channel constant over the entire slot...
channelRG(:, :, iLayer) = repmat(channelTF, 1, obj.nSymbolsSlot);
% ... but for CFO.
if cfo ~= 0
cfoVal = CPDurations * SubcarrierSpacing;
cfoVal(2:end) = cfoVal(2:end) + 1;
cfoVal = cumsum(cfoVal) * cfo;
cfoVal = exp(2j * pi * cfoVal);
channelRG(:, :, iLayer) = (channelRG(:, :, iLayer) .* cfoVal) * exp(1j * pi * ((fftSize - 1) / fftSize) * cfo);
end
% Compute received resource grid.
receivedRG = receivedRG + channelRG(:, :, iLayer) .* transmittedRG(:, :, iLayer);
end
SNR = 20; % dB
noiseVar = 10^(-SNR/10);
noise = randn(size(receivedRG)) + 1j * randn(size(receivedRG));
noise(receivedRG == 0) = 0;
noise = noise * sqrt(noiseVar / 2);
receivedRG = receivedRG + noise;
EstimatorConfig.scs = SubcarrierSpacing * 1000;
EstimatorConfig.CyclicPrefixDurations = CPDurations;
EstimatorConfig.Smoothing = configuration.smoothing;
EstimatorConfig.CFOCompensate = configuration.cfocompensate;
[channelEst, noiseEst, rsrp, epre, timeAlignment, cfoEst] = srsChannelEstimator(approxbf16(receivedRG), ...
pilots, betaDMRS, hop1, hop2, EstimatorConfig);
results = struct(...
'ChannelEst', channelEst, ...
'NoiseEst', noiseEst, ...
'RSRP', rsrp, ...
'EPRE', epre, ...
'TimeAlignment', timeAlignment, ...
'CFO', cfoEst ...
);
channel = struct(...
'RG', channelRG, ...
'Delay', channelDelay, ...
'CFO', cfo, ...
'SNR', SNR ...
);
end % of if nargout > 1
end % of function configureAndMatlab(obj, ...)
end % of methods (Access = private)
end % of classdef srsChEstimatorUnittest
function channel = configureChannel(chModel, delay, doppler, SampleRate, SubcarrierSpacing, nLayers)
channel = nrTDLChannel;
channel.NumTransmitAntennas = nLayers;
channel.NumReceiveAntennas = 1;
channel.MaximumDopplerShift = doppler;
channel.SampleRate = SampleRate;
channel.RandomStream = 'Global stream';
if strcmp(chModel, 'pure-delay')
channel.DelayProfile = 'Custom';
if isempty(delay)
% Random delay, at most one fourth of the cyclic prefix length.
channel.PathDelays = rand() * 0.018 / SubcarrierSpacing / 1000;
else
channel.PathDelays = delay;
end
channel.AveragePathGains = 0;
channel.FadingDistribution = 'Rayleigh';
elseif ismember(chModel, {'TDL-A', 'TDL-B', 'TDL-C', 'TDL-D', 'TDL-E'})
channel.DelayProfile = chModel;
if ~isempty(delay)
channel.DelaySpread = delay;
end
elseif ismember(chModel, {'TDLA30', 'TDLB100', 'TDLC300', 'TDLC60'})
channel.DelayProfile = chModel;
else
error('srsgnb_matlab:srsChEstimatorUnittest:configureChannel', ...
'Unknown channel model %s', chModel);
end
end
function mustBeConfiguration(a)
if ~isfield(a, 'nPRBs')
eidType = 'srsChEstimatorUnittest:characterize';
msgType = 'Missing configuration field "nPRBs."';
throwAsCaller(MException(eidType, msgType));
end
mustBeScalarOrEmpty(a.nPRBs);
mustBeInteger(a.nPRBs);
mustBeInRange(a.nPRBs, 1, 273);
if ~isfield(a, 'symbolAllocation')
eidType = 'srsChEstimatorUnittest:characterize';
msgType = 'Missing configuration field "symbolAllocation".';
throwAsCaller(MException(eidType, msgType));
end
mustBeVector(a.symbolAllocation)
if numel(a.symbolAllocation) ~= 2
eidType = 'srsChEstimatorUnittest:characterize';
msgType = 'Configuration field "symbolAllocation" should be an array of two elements.';
throwAsCaller(MException(eidType, msgType));
end
mustBeInteger(a.symbolAllocation);
mustBeNonnegative(a.symbolAllocation);
if (a.symbolAllocation(1) + a.symbolAllocation(2) > 14)
eidType = 'srsChEstimatorUnittest:characterize';
msgType = 'Inconsistent symbol allocation.';
throwAsCaller(MException(eidType, msgType));
end
if ~isfield(a, 'dmrsOffset')
eidType = 'srsChEstimatorUnittest:characterize';
msgType = 'Missing configuration field "dmrsOffset".';
throwAsCaller(MException(eidType, msgType));
end
mustBeScalarOrEmpty(a.dmrsOffset);
mustBeMember(a.dmrsOffset, [0, 1]);
if ~isfield(a, 'dmrsStrideSCS')
eidType = 'srsChEstimatorUnittest:characterize';
msgType = 'Missing configuration field "dmrsStrideSCS".';
throwAsCaller(MException(eidType, msgType));
end
mustBeScalarOrEmpty(a.dmrsStrideSCS);
mustBeMember(a.dmrsStrideSCS, [1, 2, 3]);
if ~isfield(a, 'dmrsStrideTime')
eidType = 'srsChEstimatorUnittest:characterize';
msgType = 'Missing configuration field "dmrsStrideTime".';
throwAsCaller(MException(eidType, msgType));
end
mustBeScalarOrEmpty(a.dmrsStrideTime);
mustBeMember(a.dmrsStrideTime, [1, 2, 4, 6, 20]);
if ~isfield(a, 'betaDMRS')
eidType = 'srsChEstimatorUnittest:characterize';
msgType = 'Missing configuration field "betaDMRS".';
throwAsCaller(MException(eidType, msgType));
end
mustBeScalarOrEmpty(a.betaDMRS);
mustBeMember(a.betaDMRS, [-3, 0]);
end
function crlb = computeCRLB(prbMask, reMask)
%computeCRLB Cramer-Rao Lower Bound
% CRLB = computeCRLB(PRBMASK, REMASK) computes the Cramer-Rao Lower Bound (CRLB)
% for the channel estimation. The CRLB is computed assuming that the entire band
% can be used for the estimation, with pilots spaced according to REMASK (first
% entry) or with pilots in all REs (second entry). The assumption is needed to
% avoid a singular Fisher matrix.
Nprb = length(prbMask);
Nre = Nprb * 12; assert(length(reMask) == 12); E = diag(kron(ones(Nprb, 1), reMask)); Jbig = ifft(fft(E, Nre, 2), Nre, 1); cp = floor(Nre / 10); J = Jbig(1:cp, 1:cp);
s = warning('error', 'MATLAB:nearlySingularMatrix');