CN100486119C - Structuring LDPC coding method - Google Patents

Structuring LDPC coding method Download PDF

Info

Publication number
CN100486119C
CN100486119C CNB2006100191625A CN200610019162A CN100486119C CN 100486119 C CN100486119 C CN 100486119C CN B2006100191625 A CNB2006100191625 A CN B2006100191625A CN 200610019162 A CN200610019162 A CN 200610019162A CN 100486119 C CN100486119 C CN 100486119C
Authority
CN
China
Prior art keywords
matrix
check
formula
input
vector
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.)
Expired - Fee Related
Application number
CNB2006100191625A
Other languages
Chinese (zh)
Other versions
CN1859012A (en
Inventor
刘文明
朱光喜
邓勇强
王玲
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Huazhong University of Science and Technology
Original Assignee
Huazhong University of Science and Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Huazhong University of Science and Technology filed Critical Huazhong University of Science and Technology
Priority to CNB2006100191625A priority Critical patent/CN100486119C/en
Publication of CN1859012A publication Critical patent/CN1859012A/en
Application granted granted Critical
Publication of CN100486119C publication Critical patent/CN100486119C/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Landscapes

  • Error Detection And Correction (AREA)
  • Compression, Expansion, Code Conversion, And Decoders (AREA)

Abstract

This invention discloses a structured LDPC coding method including: inputting parameter N,K to generate a matrix Hp, which is the matrix of the N-K stage, selecting parameters a and b to generate a linear congruence AO, A1, K, AP and KAM-1, then generating a matrix Hd, checking the matrix H based on H=[Hp|Hd], inputting an information bit sequence cd={di}, seeking the checked bit p1 and utilizing the iteration to get the rest check bit to constitute the check vectorcp={cd|cp} to get the code words c=[ cd|dp].

Description

A kind of structurized LDPC coding method
Technical field
The invention belongs to communication technical field, relate to a kind of structurized LDPC coding method, this method is the product of electronics, communication and computer technology combination.
Background technology
Since Shannon information theory was set up, channel coding theorem mainly obtained practical application in following six fields: deep space communication and satellite communication, transfer of data, storage, mobile communication, file transfer and digital audio video transmission.Five class standards have been produced thus: the Aerospace Data Systems coding standard that is used for deep space and satellite communication, the trellis coded standard that is used for the high speed data modulations demodulation, the RS coding standard that is used for disk compression storage system, the convolution coding standard of mobile cellular communication system is used for the CRC coding standard of HDLC agreement.These are used and clearly to have explained error control coding method rich in the modern digital technology is used, diversity and importance.At present, the next generation network of studying (NGN) is organically blending of all new generation network technology such as soft switchcall server net, IPv6 data the Internet, 3G (Third Generation) Moblie net, various broadband access network and intelligent optical fiber transmission network, require network to have higher data transmission rate, bigger power system capacity, high spectrum utilization more, be more suitable for various channel conditions and Geng Gao QoS, can realize collecting the transmission that speech, image, video and data are the multimedia integration business of one.Challenge in the face of the next generation network application, existing error correction coding is difficult to satisfy its transmission requirement, so a collection of new and effective error correction coding arises at the historic moment, one of error correcting code that wherein performance is best, computation complexity is minimum is exactly low-density checksum (LDPC) sign indicating number.
Since the new discovery of LDPC code weight, communication enineer and coding theory man have carried out number of research projects, and these work comprise: 1, the limiting performance analysis of LDPC sign indicating number, as density Evolution Theory, capacity limit, minimum range limit etc.; 2, efficient iterative decoding algorithm research as BP algorithm, MP algorithm, sum-product algorithm, minimum-sum algorithm and bit flipping algorithm, comprises their performance evaluation and improvement, effectively realizes design and convergence analysis; 3, the design of structural analysis and building method, as structure based on the regular and irregular sign indicating number of sparse graph, factor graph or Tanner figure, the structure of finite group multi-element code, and the design of random structure sign indicating number and Algebraic Structure sign indicating number etc.; 4, encryption algorithm design mainly is the research of linear complexity encryption algorithm; 5, the LDPC sign indicating number is in the application of different field, as the LDPC sign indicating number at the magnetic recording channel, high speed optical fiber communication, deep space and satellite communication, Digital Subscriber Line is with combining of CDMA and OFDM technology, with combining of modulation technique etc.This, different aspects multi-faceted, great dynamics to the LDPC sign indicating number studies show that the LDPC sign indicating number is with a wide range of applications in the communications field of new generation.
Along with going deep into that LDPC sign indicating number encryption algorithm is studied, some good random coded building methods have appearred, though these methods have good error-correcting performance, be not suitable for practical application.Also occurred some deterministic algebraic coding methods now, these methods can generate check matrix fast, but often need higher encoder complexity.Therefore, designing deterministic encoder with uniform enconding complexity, is that the LDPC coding techniques moves towards practical key.This patent focuses on the research of the Algebraic Structure encoder with uniform enconding complexity of LDPC sign indicating number, propose a kind of algebraically building method of practicality, coding is simple, only needs linear complexity, only need Several Parameters can generate check matrix at the decoding end, saved memory space greatly.
Summary of the invention
The objective of the invention is to overcome the deficiency of more existing building methods, a kind of structurized LDPC coding method is provided, this method has reduced encoder complexity, has saved memory space.
The invention provides a kind of structurized LDPC coding method, the steps include:
(1) constructs biconjugate angular moment battle array H respectively pAnd matrix H d
According to input parameter N and K, wherein K is an information bit length, and N is a code word size, adopts the square formation on biconjugate angular moment formation formula generation N-K rank, as matrix H p
Choose parameter a, b, a, b is an integer, and satisfies following condition (C1)-(C11), generates linear conguential sequences A according to following formula (I) 0, A 1... A P... A M-1, again according to following formula (II) generator matrix H d
(C1) a<M, b<M, wherein M=3K;
(C2) b and M prime number each other;
(C3) for the prime number p that can divide exactly M arbitrarily, a-1 is its multiple;
(C4) a and M prime number each other;
(C5) a-1 is divided exactly M;
(C6)b>2(ρ-1);
(C7)(a-1)>4b+2(ρ-1);
(C8) X 〉=ρ or (a-1)≤(M/ ρ);
(C9) there is not the satisfied (i (1≤i≤ρ-1) of (a+1) * i) mod X=0;
(C10)2ρ-1<b;
(C11)2ρ-1<2b;
A N+1=(aA n+ b) mod M, 0≤n≤M-2 formula (I)
Formula (II)
λ=3 wherein, &rho; = 3 K N - K
(2) according to H=[H p| H d] the structure check matrix H;
(3) input information bits sequence c d={ d i, i=1 ..., K, (III) asks for check bit p by following formula 1:
p 1 = &Sigma; j = 1 K h 1 j d d j Formula (III)
Wherein, h 1 j d = H d ( 1 , j ) , d j∈ c dInformation bit for input;
(4) utilize the iterative algorithm in the following formula (IV) to ask for all the other check bits, form check vector c p={ p i, i=1 ..., N-K;
p i = p i - 1 + &Sigma; j = 1 K h ij d d j I=2 ..., N-K formula (IV)
Wherein, h ij d = H d ( i , j ) , d j∈ c dInformation bit for input;
(5) according to check vector c that above-mentioned steps obtained p={ p i, and dope vector c d={ d i, obtain coding codeword c = [ c d | c p ] .
The present invention relies on platform, effectively in conjunction with electronic technology and computer technology, can realize the LDPC sign indicating number design of uniform enconding complexity.The present invention generates biconjugate angular moment battle array H respectively according to the parameter of input pAnd regular matrix H dBecause check matrix has adopted biconjugate angular moment battle array, encoder generates check vector according to the dope vector of input by alternative manner.Particularly, the present invention has the following advantages:
(1) coding has linear complexity.Owing to adopted biconjugate angular moment battle array, can generate check bit by iteration, and not need to carry out Gaussian elimination, reduced encoder complexity.
(2) save memory space.Owing to be the algebraic coding method, do not need to store check matrix at receiving terminal, only need the storage Several Parameters just can generate check matrix immediately, greatly saved memory space.
(3) coding that is obtained has preferable performance.This coded system is the coding method of quasi-regular, has effectively eliminated two wires circulation and the circulation of four lines, has excellent performance.
Description of drawings
Fig. 1 is LDPC sign indicating number applicating flow chart in wireless communication system.
Fig. 2 is the schematic flow sheet of the inventive method.
Fig. 3 is a LDPC sign indicating number decoding schematic flow sheet.
Fig. 4 is the Tanner schematic diagram (bipartite graph) of regular check matrix.
Embodiment
The present invention is described in detail below in conjunction with accompanying drawing and example.
As shown in Figure 1, the simulation flow of LDPC sign indicating number in communication system is: signal source is encoded through the LDPC code coder, the coding codeword that obtains is modulated through modulator, transmit through wireless channel, obtain soft demodulating information in the receiving terminal demodulation, decipher by ldpc decoder, recover information bit.This emulation mainly comprises: LDPC encoder, modulator, channel simulator, demodulator, ldpc decoder or the like.
At first provide the method for designing of the check matrix of LDPC sign indicating number, provide its iteration coding method then.The present invention is further illustrated below in conjunction with Fig. 2:
1 check matrix H
Check matrix constitutes H=[H by two submatrixs p| H d], introduce the building method of these two submatrixs below respectively.
1.1 biconjugate angular moment battle array H p
Input parameter N, K, wherein K is an information bit length, N is a code word size, generator matrix H p, H pBe the square formation on N-K rank, adopt biconjugate angular moment formation formula, in the following example shown in:
H p = 1 0 L 0 0 0 1 1 L 0 0 0 0 1 1 L 0 0 M M L L 0 0 0 0 L 1 1 0 0 0 L 0 1 1 - - - ( 1 )
1.2 regular matrix H d
Input parameter N, K, a, b and A 0, wherein K is an information bit length, N is a code word size, A 0Be the initial value of formula (2), can get the arbitrary value between the 0:K λ, a, b are integer.
Matrix H dCan regard the check matrix of the LDPC sign indicating number of a canonical as, for the sparse property of taking into account check matrix and the performance of encoder, the weight of choosing every row is λ=3, and then the weight of every row is &rho; = 3 K N - K . Every row are regarded as be that information node, every row are regarded as and be check-node, each information node links to each other with 3 check-nodes, and each check-node links to each other with ρ information node, and then check matrix can be described by Fig. 4.As shown in Figure 4, λ=3, ρ=6.Information node and check-node have M=3K bar access path, matrix H dCan be that the mapping one by one of M generates by a length.
H dMatrix is constructed by linear conguential sequences, at first adopts the method for iteration to generate linear conguential sequences A 0, A 1, K A p, K A M-1
A n+1=(aA n+b)mod?M,0≤n≤M-2 (2)
Then, according to sequence A 0, A 1, KA p, K A M-1By following formula generator matrix H d:
Figure C200610019162D00083
1.3 structure H dConstraints
Shine upon one by one between the connected node of information node and the connected node of check-node, its mapping must meet the following conditions:
1, an information node can not be connected to same check-node (avoiding the two wires circulation) by two paths.When
Figure C200610019162D00091
Or
Figure C200610019162D00093
The time, there is the two wires circulation.
2, do not have short circulation to exist, avoid the existence of four lines circulation (two information nodes are continuous with identical a pair of check-node).
For (2) formula, in order to be shone upon A one by one 0Can get the arbitrary integer between the 0:M-1, a and b must meet the following conditions:
C1.a<M,b<M;
C2.b and M be prime number each other;
C3. for the prime number p that can divide exactly M arbitrarily, a-1 is its multiple.In addition, if M is 4 multiple, then a-1 must be 4 multiple;
C4.a and M be prime number each other.
4 conditions above satisfying, formula (2) energy iteration generates the arbitrary integer between the 0:M-1.
If increase followingly to a, the constraints of b can be so that the pairing bipartite graph of Hd has two wires circulation and the circulation of four lines.
λ=3, H dBy iterative formula A N+1=(aA n+ b) mod M produces, a, the b C1:C4 that satisfies condition, and X=(M/ (a-1)), if the increase condition:
C5.a-1 is divided exactly M;
C6.b>2(ρ-1);
C7.(a-1)>4b+2(ρ-1);
C8.X 〉=ρ or (a-1)≤(M/ ρ);
C9. there is not the satisfied (i (1≤i≤ρ-1) of (a+1) * i) mod X=0;
H then dCorresponding bipartite graph does not have two wires circulation and the circulation of four lines.
In order on the pairing bipartite graph of check matrix H, there not to be the circulation of four lines, must increase following condition:
C10.2ρ-1<b;
C11.2ρ-1<2b。
2 iteration coding
Because the application of biconjugate angular moment battle array does not need generator matrix G just can generate check bit by iteration in cataloged procedure.Pairing code vector c is decomposed into the corresponding check vector C with the H matrix pWith dope vector c d, c=[c is promptly arranged pc d], following relation is arranged between check matrix H and the code vector c:
Hc T = ( H p H d ) c p c d = H p c p + H d c d = 0 T - - - ( 3 )
2.1 ask for check bit p 1
If input information vector C d={ d i, i=1, L, K can ask for check vector c according to (3) formula p={ p i, i=1, L, N-K is easy to try to achieve check bit p according to equation (4) 1Value:
p 1 = &Sigma; j = 1 K h 1 j d d j - - - ( 4 )
Wherein h 1 j d = H d ( 1 , j ) , d j∈c d
2.2 ask for all the other check bits
Utilize iterative algorithm, the restriction relation of establishing according to check matrix generates all the other check bits, thereby obtains coding codeword, and all the other can get by recursion:
p i = p i - 1 + &Sigma; j = 1 K h ij d d j i=2,L,N-K (5)
Wherein h ij d = H d ( i , j ) , d j∈c d。Check vector c pWith known dope vector c dMerge, obtain coding codeword c = [ c p | c d ] .
Hence one can see that: given any one dope vector c d, needn't invert to check matrix H, can iteration try to achieve c according to top method p, coding has linear complexity.
As shown in Figure 3, at the decoding end, the soft information that demodulation obtains is sent into decoder with the check matrix H that generates according to input parameter as Given information, through iterative decoding, and the information bit that obtains transmitting.
Fig. 4 is the Tanner figure of regular LDPC sign indicating number, i.e. bipartite graph.By this figure, can introduce regular matrix H intuitively dBuilding method.
In order to further specify, provide a simplified example of the present invention as follows.
Input parameter N=64, K=32, because λ=3, ρ=3 then, M=3 * 32=2 5* 3, according to top restrictive condition, a-1 can be decomposed into the form that prime factor multiplies each other, and has only two factors 2,3.
According to condition C 3, a-1 can only get 12,24,48,96.
Condition C 8 requires (a-1)≤32, and C6 requires b〉4, C7 requires (a-1)〉4b+2 (ρ-1)〉20, therefore (a-1)=24 are so a=25 gets b=5, A simultaneously 0=0.
Therefore, in the specific implementation process, input parameter is N=64, K=32, a=25, b=5, A 0=0, can obtain following matrix.
H p
1?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0
1?1?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0
0?1?1?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0
0?0?1?1?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0
0?0?0?1?1?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0
0?0?0?0?1?1?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0
0?0?0?0?0?1?1?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0
0?0?0?0?0?0?1?1?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0
0?0?0?0?0?0?0?1?1?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0
0?0?0?0?0?0?0?0?1?1?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0
0?0?0?0?0?0?0?0?0?1?1?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0
0?0?0?0?0?0?0?0?0?0?1?1?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0
0?0?0?0?0?0?0?0?0?0?0?1?1?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0
0?0?0?0?0?0?0?0?0?0?0?0?1?1?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0
0?0?0?0?0?0?0?0?0?0?0?0?0?1?1?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0
0?0?0?0?0?0?0?0?0?0?0?0?0?0?1?1?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0
0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?1?1?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0
0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?1?1?0?0?0?0?0?0?0?0?0?0?0?0?0?0
0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?1?1?0?0?0?0?0?0?0?0?0?0?0?0?0
0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?1?1?0?0?0?0?0?0?0?0?0?0?0?0
0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?1?1?0?0?0?0?0?0?0?0?0?0?0
0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?1?1?0?0?0?0?0?0?0?0?0?0
0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?1?1?0?0?0?0?0?0?0?0?0
0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?1?1?0?0?0?0?0?0?0?0
0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?1?1?0?0?0?0?0?0?0
0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?1?1?0?0?0?0?0?0
0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?1?1?0?0?0?0?0
0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?1?1?0?0?0?0
0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?1?1?0?0?0
0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?1?1?0?0
0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?1?1?0
0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?1?1
H dFor:
1?0?0?0?0?0?0?0?0?1?0?1?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0
1?0?0?0?0?1?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?1?0?0?0?0?0?0?0?0?0
0?0?1?0?0?0?0?0?0?0?0?0?0?1?0?0?0?0?0?0?0?0?0?0?0?0?0?1?0?0?0?0
0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?1?0?0?0?0?1?1?0?0?0?0?0?0?0
0?0?0?0?1?0?0?0?0?0?0?0?0?1?0?1?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0
0?0?0?0?1?0?0?0?0?1?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?1?0?0?0?0?0
0?0?0?0?0?0?1?0?0?0?0?0?0?0?0?0?0?1?0?0?0?0?0?0?0?0?0?0?0?0?0?1
0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?1?0?0?0?0?1?1?0?0?0
0?0?0?0?0?0?0?0?1?0?0?0?0?0?0?0?0?1?0?1?0?0?0?0?0?0?0?0?0?0?0?0
0?0?0?0?0?0?0?0?1?0?0?0?0?1?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?1?0
0?0?0?1?0?0?0?0?0?0?1?0?0?0?0?0?0?0?0?0?0?1?0?0?0?0?0?0?0?0?0?0
1?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?1?0?0?0?0?1
0?0?0?0?0?0?0?0?0?0?0?0?1?0?0?0?0?0?0?0?0?1?0?1?0?0?0?0?0?0?0?0
0?0?1?0?0?0?0?0?0?0?0?0?1?0?0?0?0?1?0?0?0?0?0?0?0?0?0?0?0?0?0?0
0?0?0?0?0?0?0?1?0?0?0?0?0?0?1?0?0?0?0?0?0?0?0?0?0?1?0?0?0?0?0?0
0?0?0?1?1?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?1?0
0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?1?0?0?0?0?0?0?0?0?1?0?1?0?0?0?0
0?0?0?0?0?0?1?0?0?0?0?0?0?0?0?0?1?0?0?0?0?1?0?0?0?0?0?0?0?0?0?0
0?0?0?0?0?0?0?0?0?0?0?1?0?0?0?0?0?0?1?0?0?0?0?0?0?0?0?0?0?1?0?0
0?0?1?0?0?0?0?1?1?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0
0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?1?0?0?0?0?0?0?0?0?1?0?1
0?0?0?0?0?0?0?0?0?0?1?0?0?0?0?0?0?0?0?0?1?0?0?0?0?1?0?0?0?0?0?0
0?1?0?0?0?0?0?0?0?0?0?0?0?0?0?1?0?0?0?0?0?0?1?0?0?0?0?0?0?0?0?0
0?0?0?0?0?0?1?0?0?0?0?1?1?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0
0?1?0?1?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?1?0?0?0?0?0?0?0
0?0?0?0?0?0?0?0?0?0?0?0?0?0?1?0?0?0?0?0?0?0?0?0?1?0?0?0?0?1?0?0
0?0?0?0?0?1?0?0?0?0?0?0?0?0?0?0?0?0?0?1?0?0?0?0?0?0?1?0?0?0?0?0
0?0?0?0?0?0?0?0?0?0?1?0?0?0?0?1?1?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0
0?0?0?0?0?1?0?1?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?1?0?0?0
0?1?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?0?1?0?0?0?0?0?0?0?0?0?1?0?0?0
0?0?0?0?0?0?0?0?0?1?0?0?0?0?0?0?0?0?0?0?0?0?0?1?0?0?0?0?0?0?1?0
0?0?0?0?0?0?0?0?0?0?0?0?0?0?1?0?0?0?0?1?1?0?0?0?0?0?0?0?0?0?0?0
The coding implementation procedure is as follows:
Suppose that the input information bits sequence is 01010101010101010101010101010101, according to (4) formula, can try to achieve
p 1 = &Sigma; j = 1 K h 1 j d d j = 0 ,
Remaining check bit can get by (5) formula recursion: 1100110011001100110011001100110, therefore the verification sequence that obtains of encoding is: 01100110011001100110011001100110, and coding codeword is 01,010,101,010,101,010,101,010,101,010,101,011,001,100,110,011,001,100,110 01100110.
Coding codeword transmits by channel after modulating, at the decoding end, and according to input parameter N, K, a, b and A 0, can reduce top check matrix H.Give decoder with the soft message transmission that matrix H and demodulation obtain, can obtain information sequence through iterative decoding.
Adopt the encoder of the inventive method design to have definite coding structure, only need Several Parameters just can recover check matrix, greatly saved memory space, for the practicality of LDPC sign indicating number has stepped solid step forward at the decoding end.

Claims (1)

1, a kind of structurized LDPC coding method the steps include:
(1) constructs biconjugate angular moment battle array H respectively pAnd matrix H d
According to input parameter N and K, wherein K is an information bit length, and N is a code word size, adopts the square formation on biconjugate angular moment formation formula generation N-K rank, as matrix H p
Choose parameter a, b, a, b is an integer, and satisfies following condition (C1)-(C11), generates linear conguential sequences A according to following formula (I) 0, A 1... A p... A M-1, again according to following formula (II) generator matrix H d
(C1) a<M, b<M, wherein M=3K;
(C2) b and M prime number each other;
(C3) for the prime number p that can divide exactly M arbitrarily, a-1 is its multiple;
(C4) a and M prime number each other;
(C5) a-1 is divided exactly M;
(C6)b>2(ρ-1);
(C7)(a-1)>4b+2(ρ-1);
(C8) X 〉=ρ or (a-1)≤(M/ ρ);
(C9) there are not satisfied (i of (a+1) * i) modX=0, wherein 1≤i≤ρ-1;
(C10)2ρ-1<b;
(C11)2ρ-1<2b;
A N-1=(aA n+ b) modM, 0≤n≤M-2 formula (I)
Figure C200610019162C00021
Formula (II)
λ=3 wherein, &rho; = 3 K N - K
(2) according to H=[H p| H d] the structure check matrix H;
(3) input information bits sequence c d={ d i, i=1 ..., K, (III) asks for check bit p by following formula 1:
p 1 = &Sigma; j = 1 K h 1 j d d j Formula (III)
Wherein, h 1 j d = H d ( 1 , j ) , d i∈ c dInformation bit for input;
(4) utilize the iterative algorithm in the following formula (IV) to ask for all the other check bits, form check vector c p={ p i, i=1 ..., N-K;
p i = p i - 1 + &Sigma; j = 1 K h 1 j d d j I=2 ..., N-K formula (IV)
Wherein, h ij d = H d ( i , j ) , d j∈ c dInformation bit for input;
(5) according to check vector c that above-mentioned steps obtained p={ p i, and dope vector c d={ d i, obtain coding codeword c=[c d| c p].
CNB2006100191625A 2006-05-26 2006-05-26 Structuring LDPC coding method Expired - Fee Related CN100486119C (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CNB2006100191625A CN100486119C (en) 2006-05-26 2006-05-26 Structuring LDPC coding method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CNB2006100191625A CN100486119C (en) 2006-05-26 2006-05-26 Structuring LDPC coding method

Publications (2)

Publication Number Publication Date
CN1859012A CN1859012A (en) 2006-11-08
CN100486119C true CN100486119C (en) 2009-05-06

Family

ID=37297880

Family Applications (1)

Application Number Title Priority Date Filing Date
CNB2006100191625A Expired - Fee Related CN100486119C (en) 2006-05-26 2006-05-26 Structuring LDPC coding method

Country Status (1)

Country Link
CN (1) CN100486119C (en)

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101471672B (en) * 2007-12-27 2011-04-13 华为技术有限公司 Method for generating check matrix and corresponding encoding method and encoder
JP4898858B2 (en) * 2009-03-02 2012-03-21 パナソニック株式会社 Encoder, decoder and encoding method
CN103220005B (en) * 2013-05-02 2017-04-12 广州海格通信集团股份有限公司 Method for generating LDPC (Low Density Parity Check) code check matrix and LDPC code coding method
CN104092536B (en) * 2014-05-24 2018-02-16 中国人民解放军信息工程大学 The information reconciliation method being easy in hard-wired quantum key distribution
CN106201781B (en) * 2016-07-11 2019-02-26 华侨大学 A kind of cloud date storage method based on the right canonical correcting and eleting codes
CN107786327B (en) * 2016-08-31 2020-11-13 电子科技大学 Safe and reliable transmission method based on LDPC code

Also Published As

Publication number Publication date
CN1859012A (en) 2006-11-08

Similar Documents

Publication Publication Date Title
CN101453297B (en) Encoding method and apparatus for low density generation matrix code, and decoding method and apparatus
Jouguet et al. Long-distance continuous-variable quantum key distribution with a Gaussian modulation
CN1593012B (en) Device and method for bit labeling for amplitude phase shift constellation used with low density parity check (LDPC) codes
CN101924565B (en) LDPC encoders, decoders, systems and methods
CN111030783B (en) Data transmission method and system for bit interleaving combined source-channel coding modulation
CN100505555C (en) Method for constructing low-density and irregular parity checking code
CN100486119C (en) Structuring LDPC coding method
CN101459430B (en) Encoding method and apparatus for low density generation matrix code
CN106230489A (en) It is applicable to the polarization code code modulating method of any high order modulation
CN107395319B (en) Code rate compatible polarization code coding method and system based on punching
KR20080077992A (en) Inspection matrix generation method, encoding method, communication device, communication system, and encoder
CN101588183B (en) Method and system for coding LDPC code rapidly
CN1933336B (en) Construction of LDPC (low density parity check) codes using GRS (generalized reed-solomon) code
CN107124251A (en) A kind of polarization code encoding method based on arbitrary kernel
CN103199877B (en) Method for constructing and coding structured LDPC (Low Density Parity Check) convolutional codes
Yang et al. From LDPC to chunked network codes
KR100669152B1 (en) Apparatus and method for coding of low density parity check code
Jin et al. Joint source-channel polarization with side information
CN1973440A (en) LDPC encoders, decoders, systems and methods
Chandar Sparse graph codes for compression, sensing, and secrecy
Skachek et al. Hybrid noncoherent network coding
CN103036577B (en) A kind of low-density checksum LDPC code coding circuit structure of low complex degree
Chen et al. Accumulate rateless codes and their performances over additive white Gaussian noise channel
CN103546166A (en) Check matrix construction method, encoding and decoding method and device for fountain codes
Chanayai et al. Fountain codes and their applications: Comparison and implementation for wireless applications

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant
C17 Cessation of patent right
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20090506

Termination date: 20140526