forked from NVlabs/sionna
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathtest_ofdm_channel_estimation.py
344 lines (312 loc) · 13.4 KB
/
test_ofdm_channel_estimation.py
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
#
# SPDX-FileCopyrightText: Copyright (c) 2021-2022 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
# SPDX-License-Identifier: Apache-2.0
#
try:
import sionna
except ImportError as e:
import sys
sys.path.append("../")
from sionna.mimo import StreamManagement
from sionna.ofdm import ResourceGrid, ResourceGridMapper, LSChannelEstimator, PilotPattern, KroneckerPilotPattern
from sionna.channel.tr38901 import Antenna, AntennaArray, UMi
from sionna.channel import gen_single_sector_topology as gen_topology
from sionna.channel import subcarrier_frequencies, cir_to_ofdm_channel, cir_to_time_channel
from sionna.channel import ApplyOFDMChannel
import pytest
import unittest
import numpy as np
import tensorflow as tf
gpus = tf.config.list_physical_devices('GPU')
print('Number of GPUs available :', len(gpus))
if gpus:
gpu_num = 0 # Number of the GPU to be used
try:
tf.config.set_visible_devices(gpus[gpu_num], 'GPU')
print('Only GPU number', gpu_num, 'used.')
tf.config.experimental.set_memory_growth(gpus[gpu_num], True)
except RuntimeError as e:
print(e)
def freq_int(h, i, j):
"""Linear interpolation along the second axis on a 2D resource grid
- h is [num_ofdm_symbols, num_subcarriers]
- i, j are arrays indicating the indices of nonzero pilots
"""
h_int = np.zeros_like(h)
h_int[i, j] = h[i, j]
x_0 = np.zeros_like(h)
x_1 = np.zeros_like(h)
y_0 = np.zeros_like(h)
y_1 = np.zeros_like(h)
x = np.zeros_like(h)
for a in range(h_int.shape[0]):
x[a] = np.arange(0, h_int.shape[1])
pilot_ind = np.where(h_int[a])[0]
if len(pilot_ind)==1:
x_0[a] = x_1[a] = pilot_ind[0]
y_0[a] = y_1[a] = h_int[a, pilot_ind[0]]
elif len(pilot_ind)>1:
x0 = 0
x1 = 1
for b in range(h_int.shape[1]):
x_0[a, b] = pilot_ind[x0]
x_1[a, b] = pilot_ind[x1]
y_0[a, b] = h_int[a, pilot_ind[x0]]
y_1[a, b] = h_int[a, pilot_ind[x1]]
if b==pilot_ind[x1] and x1<len(pilot_ind)-1:
x0 = x1
x1 += 1
h_int = (x-x_0)*np.divide(y_1-y_0, x_1-x_0, out=np.zeros_like(h), where=x_1-x_0!=0) + y_0
return h_int
def time_int(h, time_avg=False):
"""Linear interpolation along the first axis on a 2D resource grid
- h is [num_ofdm_symbols, num_subcarriers]
"""
x_0 = np.zeros_like(h)
x_1 = np.zeros_like(h)
y_0 = np.zeros_like(h)
y_1 = np.zeros_like(h)
x = np.repeat(np.expand_dims(np.arange(0, h.shape[0]), 1), [h.shape[1]], axis=1)
pilot_ind = np.where(np.sum(np.abs(h), axis=-1))[0]
if time_avg:
hh = np.sum(h, axis=0)/len(pilot_ind)
h[pilot_ind] = hh
if len(pilot_ind)==1:
h_int = np.repeat(h[pilot_ind], [h.shape[0]], axis=0)
return h_int
elif len(pilot_ind)>1:
x0 = 0
x1 = 1
for a in range(h.shape[0]):
x_0[a] = pilot_ind[x0]
x_1[a] = pilot_ind[x1]
y_0[a] = h[pilot_ind[x0]]
y_1[a] = h[pilot_ind[x1]]
if a==pilot_ind[x1] and x1<len(pilot_ind)-1:
x0 = x1
x1 += 1
h_int = (x-x_0)*np.divide(y_1-y_0, x_1-x_0, out=np.zeros_like(h), where=x_1-x_0!=0) + y_0
return h_int
def linear_int(h, i, j, time_avg=False):
"""Linear interpolation on a 2D resource grid
- h is [num_ofdm_symbols, num_subcarriers]
- i, j are arrays indicating the indices of nonzero pilots
"""
h_int = freq_int(h, i, j)
return time_int(h_int, time_avg)
def check_linear_interpolation(self, pilot_pattern, time_avg=False, mode="eager"):
"Simulate channel estimation with linear interpolation for a 3GPP UMi channel model"
scenario = "umi"
carrier_frequency = 3.5e9
direction = "uplink"
num_ut = pilot_pattern.num_tx
num_streams_per_tx = pilot_pattern.num_streams_per_tx
num_ofdm_symbols = pilot_pattern.num_ofdm_symbols
fft_size = pilot_pattern.num_effective_subcarriers
batch_size = 1
tf.random.set_seed(1)
ut_array = Antenna(polarization="single",
polarization_type="V",
antenna_pattern="omni",
carrier_frequency=carrier_frequency)
bs_array = AntennaArray(num_rows=1,
num_cols=4,
polarization="dual",
polarization_type="VH",
antenna_pattern="38.901",
carrier_frequency=carrier_frequency)
channel_model = UMi(carrier_frequency=carrier_frequency,
o2i_model="low",
ut_array=ut_array,
bs_array=bs_array,
direction=direction,
enable_pathloss=False,
enable_shadow_fading=False)
topology = gen_topology(batch_size, num_ut, scenario, min_ut_velocity=0, max_ut_velocity=30)
channel_model.set_topology(*topology)
rx_tx_association = np.zeros([1, num_ut])
rx_tx_association[0, :] = 1
sm = StreamManagement(rx_tx_association, num_streams_per_tx)
rg = ResourceGrid(num_ofdm_symbols=num_ofdm_symbols,
fft_size=fft_size,
subcarrier_spacing=30e3,
num_tx=num_ut,
num_streams_per_tx=num_streams_per_tx,
cyclic_prefix_length=0,
pilot_pattern=pilot_pattern)
frequencies = subcarrier_frequencies(rg.fft_size, rg.subcarrier_spacing)
channel_freq = ApplyOFDMChannel(add_awgn=False)
rg_mapper = ResourceGridMapper(rg)
if time_avg:
ls_est = LSChannelEstimator(rg, interpolation_type="lin_time_avg")
else:
ls_est = LSChannelEstimator(rg, interpolation_type="lin")
def fun():
x = tf.zeros([batch_size, num_ut, rg.num_streams_per_tx, rg.num_data_symbols], tf.complex64)
x_rg = rg_mapper(x)
a, tau = channel_model(num_time_samples=rg.num_ofdm_symbols, sampling_frequency=1/rg.ofdm_symbol_duration)
h_freq = cir_to_ofdm_channel(frequencies, a, tau, normalize=True)
y = channel_freq([x_rg, h_freq]) # noiseless channel
h_hat_lin, _n = ls_est([y, 0.])
return x_rg, h_freq, h_hat_lin
@tf.function
def fun_graph():
return fun()
@tf.function(jit_compile=True)
def fun_xla():
return fun()
if mode=="eager":
x_rg, h_freq, h_hat_lin = fun()
elif mode=="graph":
x_rg, h_freq, h_hat_lin = fun_graph()
elif mode=="xla":
x_rg, h_freq, h_hat_lin = fun_xla()
for tx in range(0, num_ut):
# Get non-zero pilot indices
i, j = np.where(np.abs(x_rg[0, tx, 0]))
h = h_freq[0,0,0,tx,0].numpy()
h_hat_lin_numpy = linear_int(h, i, j, time_avg)
self.assertTrue(np.allclose(h_hat_lin_numpy, h_hat_lin[0,0,0,tx,0].numpy()))
class TestLinearInterpolator(unittest.TestCase):
def test_sparse_pilot_pattern(self):
"One UT has two pilots, three others have just one"
num_ut = 4
num_streams_per_tx = 1
num_ofdm_symbols = 14
fft_size = 64
mask = np.zeros([num_ut, num_streams_per_tx, num_ofdm_symbols, fft_size], bool)
mask[...,[2,3,10,11],:] = True
num_pilots = np.sum(mask[0,0])
pilots = np.zeros([num_ut, num_streams_per_tx, num_pilots])
pilots[0,0,10] = 1
pilots[0,0,234] = 1
pilots[1,0,20] = 1
pilots[2,0,70] = 1
pilots[3,0,120] = 1
pilot_pattern = PilotPattern(mask, pilots)
check_linear_interpolation(self, pilot_pattern, mode="eager")
check_linear_interpolation(self, pilot_pattern, mode="graph")
check_linear_interpolation(self, pilot_pattern, mode="xla")
def test_kronecker_pilot_patterns_01(self):
num_ut = 4
num_streams_per_tx = 1
num_ofdm_symbols = 14
fft_size = 64
pilot_ofdm_symbol_indices = [2, 11]
rg = ResourceGrid(num_ofdm_symbols=num_ofdm_symbols,
fft_size=fft_size,
subcarrier_spacing=30e3,
num_tx=num_ut,
num_streams_per_tx=num_streams_per_tx,
cyclic_prefix_length=0,
pilot_pattern="kronecker",
pilot_ofdm_symbol_indices=pilot_ofdm_symbol_indices)
check_linear_interpolation(self, rg.pilot_pattern, mode="eager")
check_linear_interpolation(self, rg.pilot_pattern, mode="graph")
check_linear_interpolation(self, rg.pilot_pattern, mode="xla")
def test_kronecker_pilot_patterns_02(self):
"Only a single pilot symbol"
num_ut = 4
num_streams_per_tx = 1
num_ofdm_symbols = 14
fft_size = 64
pilot_ofdm_symbol_indices = [2]
rg = ResourceGrid(num_ofdm_symbols=num_ofdm_symbols,
fft_size=fft_size,
subcarrier_spacing=30e3,
num_tx=num_ut,
num_streams_per_tx=num_streams_per_tx,
cyclic_prefix_length=0,
pilot_pattern="kronecker",
pilot_ofdm_symbol_indices=pilot_ofdm_symbol_indices)
check_linear_interpolation(self, rg.pilot_pattern, mode="eager")
check_linear_interpolation(self, rg.pilot_pattern, mode="graph")
check_linear_interpolation(self, rg.pilot_pattern, mode="xla")
def test_kronecker_pilot_patterns_03(self):
"Only one pilot per UT"
num_ut = 16
num_streams_per_tx = 1
num_ofdm_symbols = 14
fft_size = 16
pilot_ofdm_symbol_indices = [2]
rg = ResourceGrid(num_ofdm_symbols=num_ofdm_symbols,
fft_size=fft_size,
subcarrier_spacing=30e3,
num_tx=num_ut,
num_streams_per_tx=num_streams_per_tx,
cyclic_prefix_length=0,
pilot_pattern="kronecker",
pilot_ofdm_symbol_indices=pilot_ofdm_symbol_indices)
check_linear_interpolation(self, rg.pilot_pattern, mode="eager")
check_linear_interpolation(self, rg.pilot_pattern, mode="graph")
check_linear_interpolation(self, rg.pilot_pattern, mode="xla")
def test_kronecker_pilot_patterns_04(self):
"Multi UT, multi stream"
num_ut = 4
num_streams_per_tx = 2
num_ofdm_symbols = 14
fft_size = 64
pilot_ofdm_symbol_indices = [2, 5, 8]
rg = ResourceGrid(num_ofdm_symbols=num_ofdm_symbols,
fft_size=fft_size,
subcarrier_spacing=30e3,
num_tx=num_ut,
num_streams_per_tx=num_streams_per_tx,
cyclic_prefix_length=0,
pilot_pattern="kronecker",
pilot_ofdm_symbol_indices=pilot_ofdm_symbol_indices)
check_linear_interpolation(self, rg.pilot_pattern, mode="eager")
check_linear_interpolation(self, rg.pilot_pattern, mode="graph")
check_linear_interpolation(self, rg.pilot_pattern, mode="xla")
def test_kronecker_pilot_patterns_05(self):
"Single UT, only pilots"
num_ut = 1
num_streams_per_tx = 1
num_ofdm_symbols = 5
fft_size = 64
pilot_ofdm_symbol_indices = np.arange(0, num_ofdm_symbols)
rg = ResourceGrid(num_ofdm_symbols=num_ofdm_symbols,
fft_size=fft_size,
subcarrier_spacing=30e3,
num_tx=num_ut,
num_streams_per_tx=num_streams_per_tx,
cyclic_prefix_length=0,
pilot_pattern="kronecker",
pilot_ofdm_symbol_indices=pilot_ofdm_symbol_indices)
check_linear_interpolation(self, rg.pilot_pattern, mode="eager")
check_linear_interpolation(self, rg.pilot_pattern, mode="graph")
check_linear_interpolation(self, rg.pilot_pattern, mode="xla")
def test_kronecker_pilot_patterns_06(self):
num_ut = 4
num_streams_per_tx = 1
num_ofdm_symbols = 14
fft_size = 64
pilot_ofdm_symbol_indices = [2,3,8, 11]
rg = ResourceGrid(num_ofdm_symbols=num_ofdm_symbols,
fft_size=fft_size,
subcarrier_spacing=30e3,
num_tx=num_ut,
num_streams_per_tx=num_streams_per_tx,
cyclic_prefix_length=0,
pilot_pattern="kronecker",
pilot_ofdm_symbol_indices=pilot_ofdm_symbol_indices)
check_linear_interpolation(self, rg.pilot_pattern, mode="eager")
check_linear_interpolation(self, rg.pilot_pattern, mode="graph")
check_linear_interpolation(self, rg.pilot_pattern, mode="xla")
def test_kronecker_pilot_patterns_with_time_averaging(self):
num_ut = 4
num_streams_per_tx = 1
num_ofdm_symbols = 14
fft_size = 64
pilot_ofdm_symbol_indices = [2,11]
rg = ResourceGrid(num_ofdm_symbols=num_ofdm_symbols,
fft_size=fft_size,
subcarrier_spacing=30e3,
num_tx=num_ut,
num_streams_per_tx=num_streams_per_tx,
cyclic_prefix_length=0,
pilot_pattern="kronecker",
pilot_ofdm_symbol_indices=pilot_ofdm_symbol_indices)
check_linear_interpolation(self, rg.pilot_pattern, mode="eager")
check_linear_interpolation(self, rg.pilot_pattern, mode="graph")
check_linear_interpolation(self, rg.pilot_pattern, mode="xla")