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
|
#!/usr/bin/env python
#
# Copyright 2015 Free Software Foundation, Inc.
#
# GNU Radio is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 3, or (at your option)
# any later version.
#
# GNU Radio 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
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with GNU Radio; see the file COPYING. If not, write to
# the Free Software Foundation, Inc., 51 Franklin Street,
# Boston, MA 02110-1301, USA.
#
import numpy as np
from encoder import PolarEncoder
from decoder import PolarDecoder
import matplotlib.pyplot as plt
def get_frozen_bit_position():
# frozenbitposition = np.array((0, 1, 2, 3, 4, 5, 6, 8, 9, 10, 12, 16, 17, 18, 20, 24), dtype=int)
# frozenbitposition = np.array((0, 1, 2, 3, 4, 5, 8, 9), dtype=int)
frozenbitposition = np.load('frozen_bit_positions_n256_k128_p0.11.npy').flatten()
print(frozenbitposition)
return frozenbitposition
def test_enc_dec_chain():
ntests = 100
n = 256
k = n // 2
frozenbits = np.zeros(n - k)
frozenbitposition = get_frozen_bit_position()
for i in range(ntests):
bits = np.random.randint(2, size=k)
encoder = PolarEncoder(n, k, frozenbitposition, frozenbits)
decoder = PolarDecoder(n, k, frozenbitposition, frozenbits)
encoded = encoder.encode(bits)
rx = decoder.decode(encoded)
if not is_equal(bits, rx):
raise ValueError('Test #', i, 'failed, input and output differ', bits, '!=', rx)
return
def is_equal(first, second):
if not (first == second).all():
result = first == second
for i in range(len(result)):
print '{0:4}: {1:2} == {2:1} = {3}'.format(i, first[i], second[i], result[i])
return False
return True
def exact_value(la, lb):
return np.log((np.exp(la + lb) + 1) / (np.exp(la) + np.exp(lb)))
def approx_value(la, lb):
return np.sign(la) * np.sign(lb) * np.minimum(np.abs(la), np.abs(lb))
def path_metric_exact(last_pm, llr, ui):
return last_pm + np.log(1 + np.exp(-1. * llr * (1 - 2 * ui)))
def path_metric_approx(last_pm, llr, ui):
if ui == int(.5 * (1 - np.sign(llr))):
return last_pm
return last_pm + np.abs(llr)
def calculate_path_metric_vector(metric, llrs, us):
res = np.zeros(llrs.size)
res[0] = metric(0, llrs[0], us[0])
for i in range(1, llrs.size):
res[i] = metric(res[i - 1], llrs[i], us[i])
return res
def test_1024_rate_1_code():
# effectively a Monte-Carlo simulation for channel polarization.
ntests = 10000
n = 256
k = n
transition_prob = 0.11
num_transitions = int(k * transition_prob)
frozenbits = np.zeros(n - k)
frozenbitposition = np.array((), dtype=int)
encoder = PolarEncoder(n, k, frozenbitposition, frozenbits)
decoder = PolarDecoder(n, k, frozenbitposition, frozenbits)
channel_counter = np.zeros(k)
possible_indices = np.arange(n, dtype=int)
for i in range(ntests):
bits = np.random.randint(2, size=k)
tx = encoder.encode(bits)
np.random.shuffle(possible_indices)
tx[possible_indices[0:num_transitions]] = (tx[possible_indices[0:num_transitions]] + 1) % 2
rx = tx
recv = decoder.decode(rx)
channel_counter += (bits == recv)
print channel_counter
print(np.min(channel_counter), np.max(channel_counter))
np.save('channel_counter_' + str(ntests) + '.npy', channel_counter)
def find_good_indices(res, nindices):
channel_counter = np.copy(res)
good_indices = np.zeros(channel_counter.size)
for i in range(nindices):
idx = np.argmax(channel_counter)
good_indices[idx] = 1
channel_counter[idx] = 0
return good_indices
def channel_analysis():
ntests = 10000
filename = 'channel_counter_' + str(ntests) + '.npy'
channel_counter = np.load(filename)
print(np.min(channel_counter), np.max(channel_counter))
channel_counter[0] = np.min(channel_counter)
good_indices = find_good_indices(channel_counter, channel_counter.size // 2)
info_bit_positions = np.where(good_indices > 0)
print(info_bit_positions)
frozen_bit_positions = np.delete(np.arange(channel_counter.size), info_bit_positions)
print(frozen_bit_positions)
np.save('frozen_bit_positions_n256_k128_p0.11.npy', frozen_bit_positions)
good_indices *= 2000
good_indices += 4000
plt.plot(channel_counter)
plt.plot(good_indices)
plt.show()
def main():
# n = 16
# k = 8
# frozenbits = np.zeros(n - k)
# frozenbitposition8 = np.array((0, 1, 2, 4), dtype=int)
# frozenbitposition = np.array((0, 1, 2, 3, 4, 5, 8, 9), dtype=int)
# print frozenbitposition
test_enc_dec_chain()
# test_1024_rate_1_code()
# channel_analysis()
if __name__ == '__main__':
main()
|
