CN111628786B - Adaptive minimum sum decoding method for LDPC code - Google Patents
Adaptive minimum sum decoding method for LDPC code Download PDFInfo
- Publication number
- CN111628786B CN111628786B CN202010667798.0A CN202010667798A CN111628786B CN 111628786 B CN111628786 B CN 111628786B CN 202010667798 A CN202010667798 A CN 202010667798A CN 111628786 B CN111628786 B CN 111628786B
- Authority
- CN
- China
- Prior art keywords
- decoding
- node
- side information
- check
- variable node
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 30
- 230000003044 adaptive effect Effects 0.000 title claims abstract description 12
- 230000005540 biological transmission Effects 0.000 claims abstract description 4
- 230000000694 effects Effects 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03M—CODING; DECODING; CODE CONVERSION IN GENERAL
- H03M13/00—Coding, decoding or code conversion, for error detection or error correction; Coding theory basic assumptions; Coding bounds; Error probability evaluation methods; Channel models; Simulation or testing of codes
- H03M13/03—Error detection or forward error correction by redundancy in data representation, i.e. code words containing more digits than the source words
- H03M13/05—Error detection or forward error correction by redundancy in data representation, i.e. code words containing more digits than the source words using block codes, i.e. a predetermined number of check bits joined to a predetermined number of information bits
- H03M13/11—Error detection or forward error correction by redundancy in data representation, i.e. code words containing more digits than the source words using block codes, i.e. a predetermined number of check bits joined to a predetermined number of information bits using multiple parity bits
- H03M13/1102—Codes on graphs and decoding on graphs, e.g. low-density parity check [LDPC] codes
- H03M13/1105—Decoding
Landscapes
- Physics & Mathematics (AREA)
- Probability & Statistics with Applications (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Error Detection And Correction (AREA)
Abstract
The invention discloses a self-adaptive minimum sum decoding method of LDPC codes, which comprises the following steps: 1) Setting the iteration number l=0 and the maximum iteration number I max =50; received sequence y to be decoded of length n i I=1, 2, …, n, BPSK modulation and gaussian channel transmission, and the result is recorded asi=1, 2, …, n; 2) Updating variable node side information; 3) Updating the side information of the check node; 4) Calculating posterior probability of the variable node; 5) Decoding judgment; 6) And (5) calculating a check equation. The decoding method introduces adaptive multiplicative factorsThe problem of overestimation of check node information in an MS decoding algorithm is solved, and the decoding performance is improved.
Description
Technical field:
the invention belongs to the field of electronic communication, and particularly relates to a self-adaptive minimum and decoding method of an LDPC code.
The background technology is as follows:
although the conventional belief propagation (Belief Propagation, BP) decoding algorithm has excellent decoding performance, a large number of multiplication operations are required in the conventional belief propagation (Belief Propagation, BP) decoding algorithm, which results in higher complexity and is not beneficial to implementation on hardware. With the proposal of the logarithmic domain BP (Log-Likelihood Ratio Belief Propagation, LLR-BP) decoding algorithm, the multiplication operation in the original algorithm is replaced by addition operation, and a hyperbolic tangent function is introduced, so that the operation amount is reduced, but the problem that the method is difficult to realize in engineering is not thoroughly solved. And performing approximate processing on check node updating operation of the LLR-BP algorithm to obtain a Minimum Sum (MS) decoding algorithm, wherein the algorithm greatly reduces the operation complexity, so that the algorithm becomes a decoding method for the LDPC code conventionally used on hardware. Although MS algorithms reduce complexity and facilitate implementation in hardware, their decoding performance is greatly reduced, because the range of tanh functions is (-1, 1), the range of arctanh functions is (- ≡infinity), in the LLR-BP algorithm, the variable node information is first subjected to tanh function operation, the range of the result is (-1, 1), then multiplication operation is performed multiple times, the range of the result becomes (0, 1), and finally the range is restored to (- ≡infinity) range by arctanh function. After this process, the result is less thanThe absolute value of the finally obtained check node information is smaller than min i′∈N(j)/i |L(v i′j ) And therefore, compared with an LLR-BP decoding algorithm, the problem of overestimation of check node information in an MS decoding algorithm is solved.
The invention comprises the following steps:
in order to solve the above problems, the present invention proposes an adaptive minimum sum decoding method for an LDPC code, which has the following technical scheme:
an adaptive minimum and decoding method of LDPC codes comprises the following specific steps:
1) Setting the iteration number l=0 and the maximum iteration number I max The method comprises the steps of carrying out a first treatment on the surface of the Received sequence y to be decoded of length n i I=1, 2, …, n, BPSK modulation and gaussian channel transmission, and the result is recorded asi＝1,2,…,n；
2) Updating variable node side information, wherein an ith variable node v i Pass to the j-th check node c j Side information L of (2) (l) (v ij ) The update formula is as follows:
wherein N (i)/j is the division of the jth check node c j All and ith variable node v i A set of adjacent check nodes;
3) Updating the side information of the check node, wherein the j-th check node c j Passed to the ith variable node v i Side information L of (2) (l) (c ji ) The update formula is as follows:
wherein,is an adaptive multiplicative factor, < >>z 1 The minimum value of the absolute value in the side information of the variable node is set; z 2 The next-smallest value is the absolute value in the side information of the variable node;
n (j)/i is the ith divided variable node v i All and jth check node c j A set of adjacent variable nodes;
4) Calculating posterior probability L of variable node (l) (q i )：
Where N (i) is all the nodes v of the ith variable i A set of adjacent check nodes;
5) Decoding judgment:
generation of decision codewords by hard decisionWherein, if L (l) (q i )>0, then->Otherwise
6) And (3) calculating a check equation:
if it isThe decoding is successful and the decoding result is output; if it isJudging whether the current iteration number l meets l < I max If so, l=l+1, and return to step 2); if not, decoding fails.
Preferably, the maximum number of iterations I max 50 times.
Compared with the prior art, the invention has the following beneficial effects:
the self-adaptive minimum and decoding method of the LDPC code of the invention is realized byIntroducing adaptive multiplicative factorsThe problem of overestimation of check node information in an MS decoding algorithm is solved, so that the decoding performance is improved.
Tests show that when LDPC codes with the code length of 155 are decoded, the decoding method is always superior to an MS decoding algorithm from the signal to noise ratio of 2dB to 4dB, and when the signal to noise ratio is 3.5dB, the decoding method starts to reach the performance of the LLR-BP decoding algorithm and is superior to the decoding effect of the LLR-BP decoding algorithm from 3.5dB to 4dB, the algorithm performance is greatly improved, and more outstanding decoding effect is obtained. When decoding LDPC code with code length of 576, the decoding method is always superior to MS decoding algorithm, and when from 2.3dB to 2.8dB, the performance of the decoding method is superior to LLR-BP decoding algorithm, and is improved by 0.2dB compared with LLR BP algorithm.
Description of the drawings:
FIG. 1 is a flow chart of a decoding method according to an embodiment;
FIG. 2 is a graph showing a comparison of decoding results of the conventional belief propagation decoding algorithm and the conventional minimum sum decoding algorithm according to the decoding method of the present invention;
FIG. 3 is a diagram showing a comparison of decoding results of the conventional belief propagation decoding algorithm and the conventional minimum sum decoding algorithm.
The specific embodiment is as follows:
the invention will be further described with reference to specific embodiments and corresponding drawings.
Embodiment one:
the adaptive minimum sum decoding algorithm of the LDPC code of the present invention is adopted in the present embodiment, and as shown in FIG. 1, the method comprises the steps of:
1) Setting the iteration number l=0 and the maximum iteration number I max =50; received sequence y to be decoded of length n i I=1, 2, …, n, BPSK modulation and gaussian channel transmission, and the result is recorded asi＝1,2,…,n；
2) Updating variable node side information, wherein an ith variable node v i Pass to the j-th check node c j Side information L of (2) (l) (v ij ) The update formula is as follows:
wherein N (i)/j is the division of the jth check node c j All and ith variable node v i A set of adjacent check nodes;
3) Updating the side information of the check node, aiming at the problem of overestimation of the check node information in the MS algorithm, the invention introduces the self-adaptive multiplicative factor in the updating of the side information of the check nodeJth check node c j Passed to the ith variable node v i Side information L of (2) (l) (c ji ) The update formula is as follows:
wherein N (j)/i is the ith divided variable node v i All and jth check node c j A set of adjacent variable nodes;
is an adaptive multiplicative factor; z 1 The minimum value of the absolute value in the side information of the variable node in the step 2); z 2 The next smallest value of the absolute value in the variable node side information in the step 2);
4) Calculating posterior probability L of variable node (l) (q i )：
Where N (i) is all the nodes v of the ith variable i A set of adjacent check nodes;
5) Decoding judgment:
generation of decision codewords by hard decisionWherein, if L (l) (q i )>0, then->Otherwise
6) And (3) calculating a check equation:
if it isThe decoding is successful and the decoding result is output; if it isJudging whether the current iteration number l meets l < I max If so, l=l+1, and return to step 2); if not, decoding fails.
Application example one:
the present embodiment applies the decoding method of the present invention, the conventional belief propagation decoding algorithm (LLR-BP decoding algorithm), and the conventional minimum sum decoding algorithm (MS algorithm) to decode LDPC codes with code lengths of 155 and 576 in a white gaussian noise channel environment, respectively. The maximum iteration number set in the decoding method of the invention is 50 times, the error frame number is 10000 times, and the decoding result is shown in fig. 2 and 3. According to fig. 2 and 3, when decoding an LDPC code with a code length of 155 and a code rate of 0.4, the MS decoding algorithm and the LLR-BP decoding algorithm differ by about 0.3dB, the decoding method of the present invention is always superior to the MS decoding algorithm from a signal-to-noise ratio of 2dB to 4dB, and when the signal-to-noise ratio is 3.5dB, the decoding method of the present invention starts to achieve the performance of the LLR-BP decoding algorithm, and from 3.5dB to 4dB, the decoding method of the present invention is superior to the decoding effect of the LLR-BP decoding algorithm, the algorithm performance is greatly improved, and a more outstanding decoding effect is obtained. When the LDPC code with the code length of 576 and the code rate of 0.5 is decoded, the decoding method is always superior to an MS decoding algorithm, and when the code length is from 2.3dB to 2.8dB, the performance of the decoding method is superior to an LLR-BP decoding algorithm, and is improved by 0.2dB compared with the performance of the LLR-BP algorithm.
Claims (2)
1. An adaptive minimum and decoding method for LDPC codes, characterized in that: the method comprises the following specific steps:
1) Setting the iteration number l=0 and the maximum iteration number I max The method comprises the steps of carrying out a first treatment on the surface of the Received sequence y to be decoded of length n i I=1, 2, …, n, BPSK modulation and gaussian channel transmission, and the result is recorded as
2) Updating variable node side information, wherein an ith variable node v i Pass to the j-th check node c j Side information L of (2) (l) (v ij ) The update formula is as follows:
wherein N (i)/j is the division of the jth check node c j All and ith variable node v i A set of adjacent check nodes;
3) Updating the side information of the check node, wherein the j-th check node c j Passed to the ith variable node v i Side information L of (2) (l) (c ji ) The update formula is as follows:
wherein,is an adaptive multiplicative factor, < >>z 1 The minimum value of the absolute value in the side information of the variable node is set; z 2 The next-smallest value is the absolute value in the side information of the variable node;
n (j)/i is the ith divided variable node v i All and jth check node c j A set of adjacent variable nodes;
4) Calculating posterior probability L of variable node (l) (q i )：
Where N (i) is all the nodes v of the ith variable i A set of adjacent check nodes;
5) Decoding judgment:
generation of decision codewords by hard decisionWherein, if L (l) (q i )>0, then->Otherwise->
6) And (3) calculating a check equation:
if it isThe decoding is successful and the decoding result is output; if it isJudging whether the current iteration number l meets l < I max If so, l=l+1, and return to step 2); if not, decoding fails.
2. The adaptive min-sum decoding method of an LDPC code of claim 1, wherein: the maximum iteration number I max 50 times.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010667798.0A CN111628786B (en) | 2020-07-13 | 2020-07-13 | Adaptive minimum sum decoding method for LDPC code |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010667798.0A CN111628786B (en) | 2020-07-13 | 2020-07-13 | Adaptive minimum sum decoding method for LDPC code |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111628786A CN111628786A (en) | 2020-09-04 |
| CN111628786B true CN111628786B (en) | 2023-12-22 |
Family
ID=72272952
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010667798.0A Active CN111628786B (en) | 2020-07-13 | 2020-07-13 | Adaptive minimum sum decoding method for LDPC code |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111628786B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108768409A (en) * | 2018-06-06 | 2018-11-06 | 重庆邮电大学 | A kind of LDPC interpretation methods based on normalization minimum value of optimization |
| CN110690906A (en) * | 2019-09-29 | 2020-01-14 | 中国科学院微电子研究所 | Dynamic self-correction minimum sum decoding method and decoder based on same |
| CN111342935A (en) * | 2020-03-04 | 2020-06-26 | 西安电子科技大学 | High-performance soft-decision decoding method based on QC-LDPC code |
-
2020
- 2020-07-13 CN CN202010667798.0A patent/CN111628786B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108768409A (en) * | 2018-06-06 | 2018-11-06 | 重庆邮电大学 | A kind of LDPC interpretation methods based on normalization minimum value of optimization |
| CN110690906A (en) * | 2019-09-29 | 2020-01-14 | 中国科学院微电子研究所 | Dynamic self-correction minimum sum decoding method and decoder based on same |
| CN111342935A (en) * | 2020-03-04 | 2020-06-26 | 西安电子科技大学 | High-performance soft-decision decoding method based on QC-LDPC code |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111628786A (en) | 2020-09-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7137060B2 (en) | Forward error correction apparatus and method in a high-speed data transmission system | |
| KR100804793B1 (en) | Method for updating Check Node in Low Density Parity Check Decoder | |
| CN109586730B (en) | Polarization code BP decoding algorithm based on intelligent post-processing | |
| CN108847848B (en) | BP decoding algorithm of polarization code based on information post-processing | |
| CN106803759A (en) | Polar yards of effective adaptive decoding method based on Gauss construction | |
| CN107612560B (en) | Polarization code early iteration stopping method based on partial information bit likelihood ratio | |
| CN108494412A (en) | A kind of multiple-factor amendment LDPC code interpretation method and device based on parameter Estimation | |
| CN110830049B (en) | LDPC decoding method based on density evolution improved offset minimum sum | |
| CN110784232B (en) | Space coupling LDPC code sliding window decoding method | |
| CN110417512B (en) | Joint iterative decoding method for CPM communication system | |
| CN107565978B (en) | BP decoding method based on Tanner graph edge scheduling strategy | |
| CN110336567B (en) | Joint iterative decoding method applied to G-LDPC coding cooperation | |
| CN106301517B (en) | Based on the satellite multi-beam joint-detection and interpretation method it is expected to propagate and system | |
| CN102811065B (en) | Mini-sum decoding correcting method based on linear minimum mean error estimation | |
| CN110535475A (en) | A kind of layered self-adapting normalization Min-Sum decoding algorithm | |
| CN108574492A (en) | A kind of improved LDPC code and long-pending decoding scheme | |
| CN101577607B (en) | Normalized min-sum decoding method capable of early ending iteration | |
| Teng et al. | Syndrome-enabled unsupervised learning for neural network-based polar decoder and jointly optimized blind equalizer | |
| CN111342935A (en) | High-performance soft-decision decoding method based on QC-LDPC code | |
| CN110690906B (en) | Dynamic self-correction minimum sum decoding method and decoder based on same | |
| CN108063623B (en) | Serial elimination decoding method of Polar code for reducing complexity | |
| CN111628786B (en) | Adaptive minimum sum decoding method for LDPC code | |
| CN108429605A (en) | Belief propagation interpretation method based on reliability classification | |
| US7840884B2 (en) | Turbo decoding with iterative estimation of channel parameters | |
| CN106603087B (en) | Based on the fountain codes increment decoding algorithm that can translate collection under a kind of wireless channel |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |