CN109981226A - A kind of combined dispatching decoding algorithm of joint source-channel coding system - Google Patents
A kind of combined dispatching decoding algorithm of joint source-channel coding system Download PDFInfo
- Publication number
- CN109981226A CN109981226A CN201910301989.2A CN201910301989A CN109981226A CN 109981226 A CN109981226 A CN 109981226A CN 201910301989 A CN201910301989 A CN 201910301989A CN 109981226 A CN109981226 A CN 109981226A
- Authority
- CN
- China
- Prior art keywords
- decoder
- source
- channel
- decoding
- decoding algorithm
- 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.)
- Pending
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
- H04L1/004—Arrangements for detecting or preventing errors in the information received by using forward error control
- H04L1/0045—Arrangements at the receiver end
- H04L1/0052—Realisations of complexity reduction techniques, e.g. pipelining or use of look-up tables
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
- H04L1/004—Arrangements for detecting or preventing errors in the information received by using forward error control
- H04L1/0056—Systems characterized by the type of code used
- H04L1/0057—Block codes
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Error Detection And Correction (AREA)
Abstract
A kind of combined dispatching decoding algorithm of joint source-channel coding system, is related to field of communication technology.Message initialization；Information updating；Carry out the decoding judgement of channel decoder and source decoder；Export estimated sequence of the source decoding sequence as original series.By dispatching decoding algorithm using shuffled simultaneously in channel decoder and source decoder simultaneously, accelerate the convergence rate of decoding algorithm,, there is good control action in phenomena such as uneven simultaneously for the intrinsic iterative convergence speed of system, to reduce whole decoding complexity, and have some improvement to the performance of system tool.
Description
Technical field
The present invention relates to fields of communication technology, compile more particularly, to a kind of combined signal source channel based on protograph LDPC code
The combined dispatching decoding algorithm of code system.
Background technique
With the high speed development of mobile communication technology and technology of Internet of things, the 5th third-generation mobile communication technology (5G's) is studied
Further promotion is arrived.Tradeoff in the 5th generation communication technology between reliability of communication system and validity be still one urgently
How key problem to be solved under conditions of guaranteeing information transfer rate improves transmission quality as far as possible.Information source is compiled
The purpose of code is the validity of raising system, and the purpose of channel coding is the reliability of raising system.In classical communication system
In system, message sink coding and channel coding are independently carried out, and the simplicity in realization to separate coding theorem in the communications field
It is once prevailing.But the establishment for separating coding theorem is not to be satisfied based on certain supposed premise in these supposed premises
In the case where, by message sink coding and channel combined consideration, the better performance that can be taken in practical application scene, and improve
The communication efficiency of whole system, joint source-channel coding (Joint Source-Channel Coding, abbreviation JSCC) are met the tendency of
And it gives birth to.
In many JSCC systems, low-density checksum (Low Density Parity-Check, abbreviation are introduced
LDPC) code is as double LDPC (Double LDPC, abbreviation D-LDPC) system of message sink coding and channel coding with its low-power consumption
Characteristic is well-known.A subclass of the protograph LDPC code as LDPC code has the advantages that fast coding and linear decoding, by it
Introduce D-LDPC system, thus constitute have low power consumption and low cost double protograph LDPC (Double Protograph LDPC,
DP-LDPC) system.
In traditional separation system, the decoding algorithm of LDPC code can be divided into two major classes: decoding and base based on hard decision
In the decoding of soft-decision.Decoding computation complexity based on hard decision is no doubt very low, but hard decision operation can lose big portion
The channel information divided, causes channel information utilization rate very low, and decoding error-correcting performance is poor.Therefore currently associated research is sentenced mainly for soft
Certainly decoding algorithm.Soft decision decoding algorithm is translated according to belief propagation (Belief Propagation, referred to as BP) is based primarily upon
Code algorithm.On the basis of the algorithm, according to the update sequence of message, soft-decision algorithm is broadly divided into two classes: concurrent messages are more
New algorithm, serial message more new algorithm.Wherein, serial message more new algorithm is broadly divided into two class rudimentary algorithms: based on verification section
The more new algorithm of point, and the more new algorithm based on variable node.
Current DP-JSCC system mainly uses joint belief propagation (the Joint Beilef based on Parallel Scheduling
Propagation, abbreviation Joint BP) decoding algorithm, decoding complicated high problem slow with convergence rate.For this problem,
The invention proposes combined dispatching decoding algorithm Joint Shuffled Scheduling, abbreviation Joint Shuffled.
Summary of the invention
Slow it is an object of the invention to be directed to the convergence rate of the decoding algorithm of existing DP-LDPC system, decoding is complicated
The problems such as high is spent, a kind of combined dispatching decoding algorithm of joint source-channel coding system based on protograph LDPC code is provided.
The present invention the following steps are included:
1) message initialization；
In step 1), the specific method of the message initialization can are as follows: respectively to source decoder variable node and letter
Road decoder variable node is initialized.
2) information updating；
In step 2), the specific method of the information updating can are as follows: is translated in channel decoder using shuffled scheduling
Code algorithm carries out information updating, and updates the message that source decoder is passed to from channel decoder；Meanwhile in source decoder
Information updating is carried out using shuffled scheduling decoding algorithm, and updates from source decoder and passes to disappearing for channel decoder
Breath.
3) the decoding judgement of channel decoder and source decoder is carried out；
In step 3), the specific method of the decoding judgement for carrying out channel decoder and source decoder can are as follows: if
Channel decoder and source decoder meet check equations or reach maximum number of iterations simultaneously, then stop iteration；If discontented
Foot, then continue iteration, return step 2).
4) estimated sequence of the output source decoding sequence as original series.
Compared with the existing decoding algorithm Joint BP of the system, the present invention has technical effect following prominent:
The present invention using shuffled scheduling decoding by being calculated in channel decoder and source decoder simultaneously simultaneously
Phenomena such as method accelerates the convergence rate of decoding algorithm, uneven simultaneously for the intrinsic iterative convergence speed of system has very well
Control action, to reduce whole decoding complexity, and have some improvement to the performance of system tool.
Detailed description of the invention
Fig. 1 is the joint decoder structure chart of DP-JSCC system.
Fig. 2 is phenomena such as iterative convergence speed of DP-JSCC system is uneven.
When Fig. 3 is statistical property p=0.02, Rsc=1/4, RccWhen=1/2, the channel decoder and letter of 3 kinds of algorithms of different
Convergence in mean the number of iterations performance simulation figure of the source decoder under a length of 3200 bit of information code, awgn channel.
When Fig. 4 is statistical property p=0.02, Rsc=1/4, RccWhen=1/2, the joint decoder of 3 kinds of algorithms of different is being believed
Breath code length is 3200 bits, the mean iterative number of time performance simulation figure under awgn channel.
When Fig. 5 is statistical property p=0.02, Rsc=1/4, RccWhen=1/2,3 kinds of algorithms of different are in information code a length of 3200
The AWGN performance simulation figure of bit.
When Fig. 6 is statistical property p=0.01, Rsc=1/4, RccWhen=1/2,3 kinds of algorithms of different are in information code a length of 3200
The AWGN performance simulation figure of bit.
Specific embodiment
Following embodiment will carry out detailed retouch to a specific embodiment of the invention and preferred simulation example in conjunction with attached drawing
It states.
1. illustrating in conjunction with attached drawing 1, joint decoder is counted on joint Tanner figure using Joint BP algorithm and information source
Characteristic parallel processed messages.Connection side between channel decoder and source decoder is the unique of two decoder exchange information
Approach.
2. being defined below to the 8 seed type log-likelihoods that the algorithm is related to:
1)WithRespectively indicate source decoder variable node (n=Ncc+1,...,Nsc+Ncc) and channel variable section
Point (n=1 ..., Ncc) initial likelihood value；
2)WithIt respectively indicates m-th of check-node in source decoder and passes to n-th of variable node and
N variable node passes to the log-likelihood of m-th of check-node；
3)WithIt respectively indicates m-th of check-node in channel decoder and passes to n-th of variable node and
N variable node passes to the log-likelihood of m-th of check-node；
4)WithN-th of variable node that channel decoder is sent to from source decoder is respectively indicated,
And source decoder log-likelihood is sent to from n-th of variable node of channel decoder.Meanwhile M (n) is indicated and n-th
The connected check-node of a variable node, M (n)/m indicate to remove m-th of school in the check-node being connected with n-th of variable node
Test node；N (m) shows that the variable node being connected with m-th of check-node, N (m)/n indicate the change being connected with m-th of check-node
N-th of variable node is removed in amount node.
3. the step of algorithm, is as follows:
(1) it initializes.Including variable node initialization and the initialization of other likelihood values.
(2) information updating.
In channel decoder, it is divided into two steps, the first step is shuffled algorithm information updating process, and second step is joint
Message process between decoder.
The first step, shuffled algorithm information updating process.As 1≤n≤NccAnd when m ∈ M (n), combined calculation variable
Node passes to check-node and check-node passes to the information of variable node:
Calculate the information that check-node passes to variable node:
Calculate the information that variable node passes to check-node:
Second step, message process between joint decoder:
Two steps are also divided into source decoder, the first step is shuffled algorithm information updating process, and second step is connection
Message process between conjunction decoder.
The first step, shuffled algorithm information updating process.Work as 1+Ncc≤n≤Nsc+Ncc, when m ∈ M (n), combined calculation
Variable node passes to check-node and check-node passes to the information of variable node:
Calculate the information that check-node passes to variable node:
Calculate the information that variable node passes to check-node:
Second step, message process between joint decoder::
(3) decoding judgement
Channel decoding sequence is expressed asSource decoding sequence is expressed asChannel variable node
It is expressed as with the log-likelihood of information source variable nodeWithIt is calculated respectively by following formula:
According to formula (10), (11), the value that (12) are calculated is made decisions.
IfWithMeet simultaneously, or have reached the maximum the number of iterations, then stops iteration；It is no
It then returns to (2) step and continues iterative decoding.
4th step, output sequenceAs source decoding sequence.
3. illustrating that Fig. 2 is the unique phenomena that the system is observed: iterative convergence speed non-uniform phenomenon in conjunction with attached drawing 2.Letter
The convergence in mean iterative rate of road decoder and source decoder is inconsistent, this entirety that will lead to joint decoder does not restrain,
So as to cause higher decoding complexity, more power consumption penalties are brought.
4. illustrate in conjunction with attached drawing 3, Fig. 3 be three kinds of algorithms of different channel decoder and source decoder it is a length of in information code
3200 bits, the convergence in mean the number of iterations performance simulation figure under awgn channel.Figure explanation individually introduces dispatching algorithm to letter
Source decoder can bring the improvement of its bigger constringency performance, and introduce shuffled algorithm in two decoders simultaneously, energy
Enough while the convergent decline of the Average Iteration of two decoders is brought, shows that joint shuffled algorithm can be controlled effectively
This kind of phenomenon.
5. illustrate in conjunction with attached drawing 4, when Fig. 4 is statistical property p=0.02, Rsc=1/4, RccWhen=1/2,3 kinds of algorithms of different
Mean iterative number of time performance simulation figure of the joint decoder under a length of 3200 bit of information code, awgn channel.It can be with by Fig. 4
Find out, joint shuffled is compared to two kinds of algorithms of partial shuffled and joint BP, in 2dB, respectively
The number of iterations for reducing 44%, 27% shows that scheduling decoding algorithm of the invention is improved in convergence rate.
6. illustrate in conjunction with attached drawing 5, when Fig. 5 is statistical property p=0.02, Rsc=1/4, RccWhen=1/2,3 kinds of algorithms of different
In the AWGN performance simulation figure of a length of 3200 bit of information code.As seen from Figure 5, joint shuffled is about in the bit error rate
It is 9 × 10-6When reach error floor, compared with joint BP, reduce some error floors, meanwhile, can be with by the figure
Find out, joint shuffled has better waterfall area performance.
7. illustrate in conjunction with attached drawing 6, when Fig. 6 is statistical property p=0.01, Rsc=1/4, RccWhen=1/2,3 kinds of algorithms of different
In the AWGN performance simulation figure of a length of 3200 bit of information code.As seen from Figure 6, when the bit error rate is 1 × 10-6When, phase
Two kinds of algorithms of joint BP and partial shuffled are compared to, are able to obtain the coding gain of 0.2dB and 0.1dB.
Simulation parameter is provided that the information source pattern that the present invention chooses is the R4JA code that code rate is 1/4 and code rate is 1/
2 AR4JA code, simulated channel are additive white Gaussian noise channel, are modulated using BPSK, and decoding maximum number of iterations is set as
30.Partial shuffled algorithm be source decoder introduce shuffled scheduling decoding algorithm, and channel decoder according to
So use BP decoding algorithm.
Claims (4)
1. a kind of combined dispatching decoding algorithm of joint source-channel coding system, it is characterised in that the following steps are included:
1) message initialization；
2) information updating；
3) the decoding judgement of channel decoder and source decoder is carried out；
4) estimated sequence of the output source decoding sequence as original series.
2. a kind of combined dispatching decoding algorithm of joint source-channel coding system as described in claim 1, it is characterised in that
In step 1), the message initialization method particularly includes: respectively to source decoder variable node and channel decoder variable
Node is initialized.
3. a kind of combined dispatching decoding algorithm of joint source-channel coding system as described in claim 1, it is characterised in that
In step 2), the information updating method particularly includes: disappeared in channel decoder using shuffled scheduling decoding algorithm
Breath updates, and updates the message that source decoder is passed to from channel decoder；Meanwhile shuffled is used in source decoder
It dispatches decoding algorithm and carries out information updating, and update the message for passing to channel decoder from source decoder.
4. a kind of combined dispatching decoding algorithm of joint source-channel coding system as described in claim 1, it is characterised in that
In step 3), the decoding judgement for carrying out channel decoder and source decoder method particularly includes: if channel decoder and
Source decoder meets check equations simultaneously or reaches maximum number of iterations, then stops iteration；If not satisfied, then continuing to change
Generation, return step 2).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910301989.2A CN109981226A (en) | 2019-04-16 | 2019-04-16 | A kind of combined dispatching decoding algorithm of joint source-channel coding system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910301989.2A CN109981226A (en) | 2019-04-16 | 2019-04-16 | A kind of combined dispatching decoding algorithm of joint source-channel coding system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109981226A true CN109981226A (en) | 2019-07-05 |
Family
ID=67084721
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910301989.2A Pending CN109981226A (en) | 2019-04-16 | 2019-04-16 | A kind of combined dispatching decoding algorithm of joint source-channel coding system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109981226A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111262590A (en) * | 2020-01-21 | 2020-06-09 | 中国科学院声学研究所 | Underwater acoustic communication information source and channel joint decoding method |
| CN112242851A (en) * | 2020-11-23 | 2021-01-19 | 湖南国科锐承电子科技有限公司 | Iterative data processing method in layered decoding of LDPC code and decoder system |
| CN113437979A (en) * | 2021-06-30 | 2021-09-24 | 华侨大学 | Structure optimization method and device of original pattern LDPC code based on non-uniform information source |
| CN113473149A (en) * | 2021-05-14 | 2021-10-01 | 北京邮电大学 | Semantic channel joint coding method and device for wireless image transmission |
| WO2023065955A1 (en) * | 2021-10-21 | 2023-04-27 | 华为技术有限公司 | Communication method and apparatus |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101005469A (en) * | 2006-12-28 | 2007-07-25 | 北京航空航天大学 | Joint coding and decoding device and method for source channel |
| CN101252409A (en) * | 2007-04-12 | 2008-08-27 | 中国科学院研究生院 | New algorithm of combined signal source channel decoding based on symbol level superlattice picture |
| CN107612557A (en) * | 2017-07-28 | 2018-01-19 | 华南理工大学 | A kind of modified Shuffled BP algorithms |
-
2019
- 2019-04-16 CN CN201910301989.2A patent/CN109981226A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101005469A (en) * | 2006-12-28 | 2007-07-25 | 北京航空航天大学 | Joint coding and decoding device and method for source channel |
| CN101252409A (en) * | 2007-04-12 | 2008-08-27 | 中国科学院研究生院 | New algorithm of combined signal source channel decoding based on symbol level superlattice picture |
| CN107612557A (en) * | 2017-07-28 | 2018-01-19 | 华南理工大学 | A kind of modified Shuffled BP algorithms |
Non-Patent Citations (2)
| Title |
|---|
| LU JING，HONG SHAOHUA，LV YIBO，WANG LIN: ""Hardware implementation of a joint source-channel"", 《JOURNAL OF CHONGQING UNIVERSITY OF POSTS AND TELECOMMUNICATIONS》 * |
| 洪少华，王琳: ""基于原模图LDPC码的分布式联合信源信道编码"", 《电子与信息学报》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111262590A (en) * | 2020-01-21 | 2020-06-09 | 中国科学院声学研究所 | Underwater acoustic communication information source and channel joint decoding method |
| CN111262590B (en) * | 2020-01-21 | 2020-11-06 | 中国科学院声学研究所 | Underwater acoustic communication information source and channel joint decoding method |
| CN112242851A (en) * | 2020-11-23 | 2021-01-19 | 湖南国科锐承电子科技有限公司 | Iterative data processing method in layered decoding of LDPC code and decoder system |
| CN112242851B (en) * | 2020-11-23 | 2024-03-19 | 湖南国科锐承电子科技有限公司 | Iterative data processing method and decoder system in layered decoding of LDPC code |
| CN113473149A (en) * | 2021-05-14 | 2021-10-01 | 北京邮电大学 | Semantic channel joint coding method and device for wireless image transmission |
| CN113437979A (en) * | 2021-06-30 | 2021-09-24 | 华侨大学 | Structure optimization method and device of original pattern LDPC code based on non-uniform information source |
| CN113437979B (en) * | 2021-06-30 | 2023-05-16 | 华侨大学 | Method and device for optimizing structure of orthographic LDPC code based on non-uniform information source |
| WO2023065955A1 (en) * | 2021-10-21 | 2023-04-27 | 华为技术有限公司 | Communication method and apparatus |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109981226A (en) | A kind of combined dispatching decoding algorithm of joint source-channel coding system | |
| Abbas et al. | High-throughput and energy-efficient belief propagation polar code decoder | |
| He et al. | A joint source and channel coding scheme base on simple protograph structured codes | |
| CN107565978B (en) | BP decoding method based on Tanner graph edge scheduling strategy | |
| CN110535475B (en) | Hierarchical adaptive normalized minimum sum decoding algorithm | |
| TWI731696B (en) | A method of decoding the polar codes based on belief propagation | |
| CN108039891A (en) | A kind of polarization code BP interpretation methods and device based on multistage more new technological process | |
| CN105577193A (en) | Loop-break based mixed weighted bit-flipping LDPC decoding method | |
| Xu et al. | Joint shuffled scheduling decoding algorithm for DP-LDPC codes-based JSCC systems | |
| Hu et al. | A comparative study of polar code decoding algorithms | |
| Deng et al. | Reduced-complexity deep neural network-aided channel code decoder: A case study for BCH decoder | |
| Levin et al. | Lazy scheduling forLDPC decoding | |
| CN107404323A (en) | A kind of LDPC code based on staggeredly ranks message transmission improves decoding algorithm | |
| CN106027069A (en) | Cyclic switching hybrid weighted bit-flipping LDPC decoding method | |
| CN108566211B (en) | Layered LDPC decoding method based on dynamic change of H matrix layer processing sequence | |
| KR20090012189A (en) | Apparatus and method for decoding using performance enhancement algorithm for ldpc codes with scaling based min-sum iterative decoding | |
| CN107612557B (en) | Improved Shuffled BP algorithm | |
| Qingle et al. | A low complexity model-driven deep learning ldpc decoding algorithm | |
| CN111669184B (en) | Joint edge dynamic scheduling iterative decoding method applied to LDPC coding cooperation | |
| CN210075203U (en) | Decoding device of LDPC code | |
| Meng et al. | LLR processing of polar codes in concatenation systems | |
| Wang et al. | Reduced latency iterative decoding of LDPC codes | |
| Huang et al. | Decoding algorithm of LDPC codes based on genetic algorithm | |
| CN103701475A (en) | Decoding method for Turbo codes with word length of eight bits in mobile communication system | |
| Honghao et al. | A Low Complexity Decoding Algorithm Design based on Quasi-Cyclic LDPC Codes |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190705 |