-
Notifications
You must be signed in to change notification settings - Fork 93
/
decoding.html
342 lines (295 loc) · 16.3 KB
/
decoding.html
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
<HTML><HEAD>
<TITLE> Decoding Received Blocks </TITLE>
</HEAD><BODY>
<H1> Decoding Received Blocks </H1>
Transmitted codewords are decoded from the received data on the basis
of the <I>likelihood</I> of the possible codewords, which is the
probability of receiving the data that was actually received if the
codeword is question were the one that was sent. This software
presently deals only with memoryless channels, in which the noise is
independent from bit to bit. For such a channel, the likelihood
factorizes into a product of likelihoods for each bit.
For decoding purposes, all that matters is the relative likelihood
for a bit to be 1 versus 0. This is captured by the <I>likelihood
ratio</I> in favour of a 1, which is P(data | bit is 1) / P(data |
bit is 0).
<P>For a Binary Symmetric Channel with error probability <I>p</I>,
the likelihood ratio in favour of a 1 bit is as follows:
<BLOCKQUOTE>
If the received data was +1: (1-<I>p</I>) / <I>p</I><BR>
If the received data was -1: <I>p</I> / (1-<I>p</I>)
</BLOCKQUOTE>
For an Additive White Gaussian Noise channel, with signals of +1 for a 1 bit
and or -1 for a 0 bit, and with noise standard deviation <I>s</I>, the
likelihood ratio in favour of a 1 bit when data <I>y</I> was received is
<BLOCKQUOTE>
exp ( 2y / s<SUP><SMALL>2</SMALL></SUP> )
</BLOCKQUOTE>
For an Additive White Logistic Noise channel, the corresponding
likelihood ratio is
<I>d</I><SUB><SMALL>1</SMALL></SUB>/<I>d</I><SUB><SMALL>0</SMALL></SUB>,
where
<I>d</I><SUB><SMALL>1</SMALL></SUB>=<I>e</I><SUB><SMALL>1</SMALL></SUB>
/ (1+<I>e</I><SUB><SMALL>1</SMALL></SUB>)<SUP><SMALL>2</SMALL></SUP> and
<I>d</I><SUB><SMALL>0</SMALL></SUB>=<I>e</I><SUB><SMALL>0</SMALL></SUB>
/ (1+<I>e</I><SUB><SMALL>0</SMALL></SUB>)<SUP><SMALL>2</SMALL></SUP>,
with <I>e</I><SUB><SMALL>1</SMALL></SUB>=exp(-(<I>y</I>-1)/<I>w</I>) and
<I>e</I><SUB><SMALL>0</SMALL></SUB>=exp(-(<I>y</I>+1)/<I>w</I>).
<BLOCKQUOTE> </BLOCKQUOTE>
<P>It is usual to consider codewords to be equally likely <I>a
priori</I>. This is reasonable if the source messages are all equally
likely (any source redundancy being ignored, or remove by a
preliminary data compression stage), provided that the mapping from
source messages to codewords is onto. Decoding can then be done using
only the parity check matrix defining the codewords, without reference
to the generator matrix defining the mapping from source messages to
codewords. Note that the condition that this mapping be onto isn't
true with this software in the atypical case where the code is defined
by a parity check matrix with redundant rows; see the discussion of <A
HREF="dep-H.html">linear dependence in parity check matrices</A>.
This minor complication is mostly ignored here, except by the exhaustive
enumeration decoding methods.
<P>Assuming equal <I>a priori</I> probabilities for codewords, the
probability of correctly decoding an entire codeword is minimized by
picking the codeword with the highest likelihood. One might instead
wish to decode each bit to the value that is most probable. This
minimizes the bit error rate, but is not in general guaranteed to lead
a decoding for each block to the most probable complete codeword;
indeed, the decoding may not be a codeword at all. Minimizing the bit
error rate seems nevertheless to be the most sensible objective,
unless block boundaries have some significance in a wider context.
<P>Optimal decoding by either criterion is infeasible for general
linear codes when messages are more than about 20 or 30 bits in
length. The fundamental advantage of Low Density Parity Check codes
is that good (though not optimal) decodings can be obtained by methods
such as probability propagation, described next.
<A NAME="prprp"><H2>Decoding by probability propagation</H2></A>
<P>The probability propagation algorithm was originally devised by
Robert Gallager in the early 1960's and later reinvented by David
MacKay and myself. It can be seen as an instance of the sum-product
algorithm for inference on factor graphs, and as an instance of belief
propagation in probabilistic networks. See the <A
HREF="refs.html">references</A> for details. Below, I give a fairly
intuitive description of the algorithm.
<P>The algorithm uses only the parity check matrix for the code, whose
columns correspond to codeword bits, and whose rows correspond to
parity checks, and the likelihood ratios for the bits derived from the
data. It aims to find the probability of each bit of the transmitted
codeword being 1, though the results of the algorithm are in general
only approximate.
<P>The begin, information about each bit of the codeword derived from
the received data for that bit alone is expressed as a <I>probability
ratio</I>, the probability of the bit being 1 divided by the
probability of the bit being 0. This probability ratio is equal to
the likelihood ratio (see above) for that bit, since 0 and 1 are
assumed to be equally likely <I>a priori</I>. As the algorithm
progresses, these probability ratios will be modified to take account
of information obtained from other bits, in conjunction with the
requirement that the parity checks be satisfied. To avoid double
counting of information, for every bit, the algorithm maintains a
separate probability ratio for each parity check that that bit
participates in, giving the probability for that bit to be 1 versus 0
based only on information derived from <I>other</I> parity checks,
along with the data received for the bit.
<P>For each parity check, the algorithm maintains separate
<I>likelihood ratios</I> (analogous to, but distinct from, the
likelihood ratios based on received data), for every bit that
participates in that parity check. These ratios give the probability
of that parity check being satisfied if the bit in question is 1
divided by the probability of the check being satisfied if the bit is
0, taking account of the probabilities of each of the <I>other</I>
bits participating in this check being 1, as derived from the
probability ratios for these bits with respect to this check.
<P>The algorithm alternates between recalculating the likelihood
ratios for each check, which are stored in the <B>lr</B> fields of the
parity check matrix entries, and recalculating the probability ratios
for each bit, which are stored in the <B>pr</B> fields of the entries
in the sparse matrix representation of the parity check matrix. (See
the documentation on <A HREF="mod2sparse.html#rep">representation of
sparse matrices</A> for details on these entries.)
<P>Recalculating the likelihood ratio for a check with respect to some
bit may appear time consuming, requiring that all possible
combinations of values for the other bits participating in the check
be considered. Fortunately, there is a short cut. One can calculate
<BLOCKQUOTE>
<I>t</I>
= product of [ 1 / (1+<I>p<SUB><SMALL>i</SMALL></SUB></I>)
- <I>p<SUB><SMALL>i</SMALL></SUB></I> /
(1+<I>p<SUB><SMALL>i</SMALL></SUB></I>) ]
= product of [ 2 / (1+<I>p<SUB><SMALL>i</SMALL></SUB></I>) - 1 ]
</BLOCKQUOTE>
where the product is over the probability ratios
<I>p<SUB><SMALL>i</SMALL></SUB></I> for the other bits participating
in this check. Factor <I>i</I> in this product is equal to probability
of bit <I>i</I> being 0 minus the probability that it is 1. The terms
in the expansion of this product (in the first form above) correspond to
possible combinations of values for the other bits, with the result that
<I>t</I> will be the probability of the check being satisfied if the bit
in question is 0 minus the probability if the bit in question is 1. The
likelihood ratio for this check with respect to the bit in question can then
be calculated as (1-<I>t</I>)/(1+<I>t</I>).
<P>For a particular check, the product above differs for different
bits, with respect to which we wish to calculate a likelihood ratio,
only in that for each bit the factor corresponding to that bit is left
out. We can calculate all these products easily by ordering the bits
arbitrarily, computing running products of the factor for the first
bit, the factors for the first two bits, etc., and also running
products of the factor for the last bit, the factors for the last two
bits, etc. Multiplying the running product of the factors up to
<I>i</I>-1 by the running product of the factors from <I>i</I>+1 on
gives the product needed for bit <I>i</I>. The second form of the
factors above is used, as it requires less computation, and is still
well defined even if some ratios are infinite.
<P>To recalculate the probability ratio for a bit with respect to a
check, all that is need is to multiply together the likelihood ratio
for this bit derived from the received data (see above), and the
current values of the likelihood ratios for all the <I>other</I>
checks that this bit participates in, with respect to this bit. To
save time, these products are computed by combining forward and
backward products, similarly to the method used for likelihood ratios.
<P>By including likelihood ratios from all checks, a similar
calculation produces the current probability ratio for the bit to be 1
versus 0 based on all information that has propagated to the bit so
far. This ratio can be thresholded at one to produce the current best
guess as to whether this bit is a 1 or a 0.
<P>The hope is that this algorithm will eventually converge to a state
where these bit probabilities give a near-optimal decoding. This is
does not always occur, but the algorithm behaves well enough to
produce very good results at rates approaching (though not yet
reaching) the theoretical Shannon limit.
<P><A NAME="decode"><HR><B>decode</B>: Decode blocks of received data
into codewords.
<BLOCKQUOTE><PRE>
decode [ -f ] [ -t | -T ] <I>pchk-file received-file decoded-file</I> [ <I>bp-file</I> ] <I>channel method</I>
</PRE>
<BLOCKQUOTE>
where <TT><I>channel</I></TT> is one of:
<BLOCKQUOTE><PRE>
bsc <I>error-probability</I>
awgn <I>standard-deviation</I>
awln <I>width</I>
</PRE></BLOCKQUOTE>
and <TT><I>method</I></TT> is one of:
<BLOCKQUOTE><PRE>
enum-block <TT><I>gen-file</I></TT>
enum-bit <TT><I>gen-file</I></TT>
prprp <TT>[-]<I>max-iterations</I></TT>
</PRE></BLOCKQUOTE>
</BLOCKQUOTE>
</BLOCKQUOTE>
<P>Decodes the blocks in <TT><I>received-file</I></TT>, which are
assumed to be have been received through the specified channel. The
results written to <TT><I>decoded-file</I></TT> are the specified
decoding method's guesses as to what bits were sent through the
channel, given what was received. The probability of each bit being a
1, as judged by the decoding method being used, is written to
<TT><I>bp-file</I></TT>, if given.
<P>A newline is output at the end of each block written to
<TT><I>decoded-file</I></TT> and <TT><I>bp-file</I></TT>. Newlines in
<TT><I>received-file</I></TT> are ignored. A warning is displayed on
standard error if the number of bits in <TT><I>received-file</I></TT>
is not a multiple of the block length.
<P>A summary is displayed on standard error, giving the total number
of blocks decoded, the number of blocks that decoded to valid
codewords, the average number of iterations of the decoding algorithm
used, and the percent of bits that were changed from the values one
would guess for them based just on their individual likelihood ratios.
<P>If the <B>-t</B> option is given, a line of information regarding each block
decoded is written to standard output, preceded by a line of headers.
The information for each block is as follows:
<BLOCKQUOTE>
<TABLE>
<tr align="left" valign="top">
<td> <B>block</B> </td>
<td>The number of the block, from zero</td></tr>
<tr align="left" valign="top">
<td> <B>iterations</B> </td>
<td>The number of "iterations" used in decoding. What exactly an iteration
is depends on the decoding method used (see
<A HREF="decode-detail.html">here</A>).</td></tr>
<tr align="left" valign="top">
<td> <B>valid</B> </td>
<td>Has the value 1 if the decoding is a valid codeword, 0 if not.</td></tr>
<tr align="left" valign="top">
<td> <B>changed</B> </td>
<td>The number of bits in the decoding that differ from the bit that would
be chosen based just on the likelihood ratio for that bit. Bits whose
likelihood ratios are exactly one contribute 0.5 to this count.</td></tr>
</TABLE>
</BLOCKQUOTE>
The file produced is is suitable for
reading into the S-Plus or R statistics packages, with a command such as
<BLOCKQUOTE><PRE>
data <- read.table(<I>file</I>,header=T)
</PRE></BLOCKQUOTE>
<P>If instead the <B>-T</B> option is given, detailed information on
the process of decoding each block will be written to standard output.
For a description, see the <A HREF="decode-detail.html">documentation
on detailed decoding trace information</A>.
<P>The type of channel that is assumed is specified after the file
name arguments. This may currently be either <TT>bsc</TT> (or
<TT>BSC</TT>) for the Binary Symmetric Channel, or <TT>awgn</TT> (or
<TT>AWGN</TT>) for the Additive White Gaussian Noise channel, or
<TT>awln</TT> (or <TT>AWLN</TT>) for the Additive White Logistic Noise
channel. The channel type is followed by an argument specifying the
assumed characteristics of the channel, as follows:
<BLOCKQUOTE>
<P>BSC: The probability that a bit will be flipped by noise - ie, the
probability that the bit received is an error.
<P>AWGN: The standard deviation of the Gaussian noise added to the
encodings of the bits.
<P>AWLN: The width parameter of the logistic distribution for the noise
that is added to the encodings of the bits.
</BLOCKQUOTE>
See the description of <A HREF="channel.html">channel transmission</A>
for more about these channels.
<P>Following the channel specification is a specification of the
decoding method to use. The <TT>enum-block</TT> and <TT>enum-bit</TT>
methods find the optimal decoding by exhaustive enumeration of
codewords derived from all possible source messages. They differ in
that <TT>enum-block</TT> decodes to the most likely codeword, whereas
<TT>enum-bit</TT> decodes to the bits that are individually most
probable. These methods require that a file containing a
representation of a generator matrix be given, to allow enumeration of
codewords. If the parity check matrix has no redundant rows, any
valid generator matrix will give the same decoding (except perhaps if
there is a tie). If redundant rows exist, the generator matrix should
specify the same set of message bits as the generator matrix that was
used for the actual encoding, since the redundancy will lead to some
codeword bits being fixed at zero (see <A HREF="dep-H.html">linear
dependence in parity check matrices</A>).
<P>The <TT>prprp</TT> decoding method decodes using <A
HREF="#prprp">probability propagation</A>. The maximum number of
iterations of probability propagation to do is given following
<TT>prprp</TT>. If a minus sign precedes this number, the maximum
number of iterations is always done. If no minus sign is present, the
algorithm stops once the tentative decoding, based on bit-by-bit
probabilities, is a valid codeword. Note that continuing to the
maximum number of iterations will usually result in
at least slightly different bit probabilities (written to
<TT><I>bp-file</I></TT> if specified), and could conceivably change
the decoding compared to stopping at the first valid codeword, or
result in a failure to decode to a valid codeword even though one was
found earlier.
<P>If the <B>-f</B> option is given, output to <TT><I>decoded-file</I></TT>
is flushed after each block. This allows one to use decode as a server,
reading blocks to decode from a named pipe, and writing the decoded block
to another named pipe.
<P><A NAME="extract"><HR><B>extract</B>: Extract the message bits from a block.
<BLOCKQUOTE><PRE>
extract <I>gen-file decoded-file extracted-file</I>
</PRE></BLOCKQUOTE>
<P>Given a file of codewords in <TT><I>decoded-file</I></TT> (usually,
decoded blocks output by <A HREF="#decode"><TT>decode</TT></A>), and a
generator matrix from <TT><I>gen-file</I></TT> (needed only to
determine where the message bits are located in a codeword), this
program writes the message bits extracted from these codewords to the
file <TT><I>extracted-file</I></TT>.
<P>A newline is output at the end of each block written to
<TT><I>extracted-file</I></TT>. Newlines in
<TT><I>decoded-file</I></TT> are ignored. A warning is displayed on
standard error if the number of bits in <TT><I>decoded-file</I></TT>
is not a multiple of the block length.
<HR>
<A HREF="index.html">Back to index for LDPC software</A>
</BODY></HTML>