WO2012037749A1

WO2012037749A1 - Encoding method for low density parity check code and apparatus thereof

Info

Publication number: WO2012037749A1
Application number: PCT/CN2010/079103
Authority: WO
Inventors: 龚贤卫; 袁志锋; 徐俊; 李长兴
Original assignee: 中兴通讯股份有限公司
Priority date: 2010-09-25
Filing date: 2010-11-25
Publication date: 2012-03-29
Also published as: CN102412842A; CN102412842B

Abstract

An encoding method for Low Density Parity Check (LDPC) code and the apparatus thereof are provided by the present invention. The method includes: comparing the code rate of a mother code supported by an unified base matrix with a transmission code rate and determining the length K' of input information packet bits based on the result of comparison; determining an encoding matrix of the LDPC code based on the unified base matrix and an expanding factor; determining the information bits to be encoded according to the received information packet bits with the length of K'; by using the encoding matrix, implementing the LDPC encoding for the information bits to be encoded and processing the encoded codeword bits of the mother code to obtain the codeword bits to be transmitted. The encoding method for LDPC code and the apparatus thereof provided by the present invention determine the base matrix and the expanding factor, determine the length of the input information packet or process the encoding matrix based on the received transmission code rate, obtain the codeword meeting the preceding transmission code rate, realize the variability of the code rate on the basis of optimal performance, reduce the hardware design complexity, and has a low cost.

Description

Method and device for encoding low density parity check code

The present invention relates to the field of digital communications, and in particular, to a method and apparatus for encoding a Low Density Parity Check (LDPC) code. Background technique

Currently, digital communication systems are commonly used communication systems. As shown in FIG. 1, the digital communication system is composed of a transmitting end, a channel, and a receiving end, where the transmitting end usually includes a source, a source encoder, a channel encoder, and a modulator (or a writing unit); Typically, a demodulator (or readout unit), a channel decoder, a source decoder, and a sink are included; there is a channel (or storage medium) between the transmitting end and the receiving end, and a noise source is present in the channel. Among them, the channel coding link (including channel coding and decoding, modulation and demodulation, etc.) is the most critical technology of the entire digital communication physical layer, which determines the effectiveness and reliability of the underlying transmission of the digital communication system.

The channel encoder is designed to combat a variety of noises and interferences during transmission. By artificially adding redundant information, the system has the ability to automatically correct errors, thereby ensuring the reliability of information transmission. LDPC code is a kind of linear block code that can be defined by a very sparse parity check matrix or bipartite graph. It is the sparsity of its check matrix that can realize low complexity compiling code, which makes LDPC practical. .

An MxN parity check matrix H defines a constraint that each N-bit codeword satisfies M parity sets, that is, the encoded N-bit codeword has an M-bit check bit, N - M Bit information bit. Let the parity check matrix H of the LDPC code be a matrix of (3⁄4 x _z )x( xz), which is composed of 3⁄4 X block matrices, each of which is a different power of the basic permutation matrix of zxz When the basic permutation matrix is an identity matrix, they are all cyclic shift matrices of the identity matrix, as shifted to the right, with the following form:

Through such a power, each block matrix can be uniquely identified. The power of the unit matrix can be represented by 0, and the power of the zero matrix is represented by -1. Thus, if each block matrix of H is replaced by its power, a power matrix is obtained. The definition here is the basic matrix of H, and the extension matrix called H. In actual coding, z = code length / number of columns of the base matrix = N / n _b , called the spreading factor. For example, matrix

H

Can be obtained by the following parameter _z , a 2x4 base matrix and a basic permutation matrix, ie a zxz unit matrix extension,

0 1 0 -1

z=3 and =

2 1 2 1 Therefore, it can also be said that the encoder of the LDPC code is uniquely generated from the fundamental matrix, the spreading factor z, and the basic permutation matrix.

If LDPC uses a basic matrix for each different spreading factor, then for each different code length, the codec of the LDPC code needs to store a basic matrix. When the code length is very large, it needs to store a lot. The basic matrix needs to consider storage and other issues. Therefore, when the variable code length needs to be realized, the LDPC code of a plurality of code lengths within a certain range of the same code rate will use the same matrix of the basic matrix. We refer to this matrix as a unified base matrix. When different code lengths,

H, for correction and extension, can get the parity check matrix H, so that the generated compiled code Can be applied to occasions where the code length is variable. Wherein, the correction is to use a spreading factor of different code lengths to correct non-negative values in the base matrix or the unified base matrix HTM such that the corrected element value is smaller than the spreading factor value under the code length. This is because or the elements in the TM represent the power of the basic permutation matrix, and the basic permutation matrix, that is, the unit matrix of zxz, once per cycle, is equivalent to shifting the unit matrix to the right by one bit, that is, the unit matrix is shifted to the right by one column. Therefore, the basic cyclic matrix is cyclically shifted by at most z-1 times; if it is repeated, the obtained unit matrix is equal to no cyclic shift. There are many correction algorithms. For example, you can use modulo (mod), round (floor) or round (round+round), etc., and set ^ as the base matrix H, the non-i-th row and the j-th column. The negative element, which is the non-negative element of the i-th row and the j-th column in the corrected base moment, has: For the modulo (mod) algorithm: three Ρ mod z three P _H for rounding (floor+floor) algorithm: P _f :

N

For rounding (scale+round) algorithm: .' = Round E x— = Round where Z is the expansion factor corresponding to the current code length, ie the number of rows or columns of the block square matrix, which is the maximum support code length N _max Corresponding expansion factor. Mod is the modulo operation, and the round is rounded off. For example, for an LDPC code with a code length of 1152 bits, set a non-negative element of the base matrix to 93, and set its maximum supported code length to 2304, and the size of the base matrix to be 12x24. The result of the correction is: Modulo (mod) algorithm: = 93 mod 48 = 45;

1152

For the rounding/floor algorithm: 93 X 6.5

2304 丄 4 ” = 46 ; For rounding (round+round) algorithm: Round(93x) = Round(46.5) = 47.

2304

Since the LDPC code of the specific code rate variable code length has the same form of the basic matrix, it is entirely possible to use an encoder/decoder, and the same form of the basic matrix can also be called a unified basic matrix.

Assuming that the size of the parity check matrix H of an LDPC code is m _b xz)x (n _b xz) and the size of the unified base matrix is x, the mother code rate supported by the LDPC code is R _m = (n _b - _mb / n _b ; If the input information packet length is K, after a series of processing to obtain a codeword bit to be transmitted of length Ν, the transmission code rate R^ K 'IN ' is called.

In IEEE802.16e, the code rate supported by the LDPC code is 1/2, 2/3, 3/4, 5/6. In the existing LDPC code coding scheme, the LDPC coding matrix with different mother code rate is usually used. H, to ensure that the LDPC code has a certain degree of flexible bit rate and performance, at this time, there will be four basic matrices requiring multiple LDPC codecs in hardware, which will greatly increase the hardware implementation cost; or use an adaptable A variety of mother code rate encoders/decoders are used to implement different code rate codecs, but the hardware design can be very complicated, which also increases the cost of hardware implementation. Summary of the invention

In view of the above, it is a primary object of the present invention to provide an encoding method and apparatus for an LDPC code that implements encoding of an LDPC code of any variable rate.

In order to achieve the above object, the technical solution of the present invention is achieved as follows:

A method for encoding a low-density parity check LDPC code, the method comprising the steps of: comparing a mother code rate and a transmission code rate supported by a unified base matrix, and determining a length K of the input information packet bit according to the comparison result ,;

Determining an encoding matrix of the LDPC code according to the unified basic matrix and the spreading factor;

Determining information bits to be encoded by receiving information packet bits of length K, performing LDPC coding on the information bits to be encoded by using the coding matrix, and encoding the information bits The obtained mother code code word bits are processed to obtain code word bits to be transmitted.

The method further includes: setting a unified base matrix and an expansion factor of the LDPC code.

The comparing the mother code rate and the transmission code rate supported by the unified basic matrix further includes: receiving the current transmission code rate;

The determining the length K of the input information packet bit according to the comparison result is:

When ? _tt ≥ ? _m , determine the length of the information packet 〖' is ( _^ ₂ ; when ? _tt <^, determine the length of the information packet bit is K' = func(n _b x _Z xR _tx /(\- R ) , or K' = func(m _b Xz/(lR _tx ))-m _b Xz , or = func(func(m _b z /(\- R _tx ))xR _tx );

Wherein, for the transmission rate, R _m is the mother code rate, m _b , respectively is the number of rows and columns of the unified base matrix, Z is the expansion factor, and j _C (x) is the integer from the X up, down An integer or rounding takes one of the integers.

Determining, according to the unified basic matrix and the spreading factor, the coding matrix of the LDPC code is: modifying the unified basic matrix, and expanding the modified basic matrix by using the unit matrix of zxz according to the spreading factor z to obtain an encoding matrix of the LDPC code .

The information packet bit of the length K is received, and the information bit to be encoded is determined as: At that time, the received information packet length of length K is directly used as the information bit to be encoded; A padding bit is added to the information packet bit to form an information bit to be encoded of length .

The processing of the encoded mother code codeword bits is performed as follows:

The mother code codeword obtained by the coding is rearranged, wherein the mother code codeword bits before the rearrangement are, Α, -, Α^, and the rearranged mother code codeword bits are n.., ^ rearrangement formula is :

^! S PV denotes a codeword rearrangement vector whose element set is m ₆ +l, - +2 ···, -1} , PV(x) denotes an element of the PV vector index x, and X is a non-negative integer; When R ≥ R _m , the rearranged mother code codeword bits A^C^-^ are taken f_[/] bits from the smallest to the largest, as the codeword bits to be transmitted;

When ^R <R _m , the y^xz- padding bits to be added are rearranged from the mother code codeword bits. The system bits in A, · · ·, are deleted, and the remaining mother code codeword bits are used as codeword bits to be transmitted.

The encoding matrix for determining the LDPC code is:

Correcting the unified basic matrix, according to the spreading factor z, using the unit matrix of z x z to extend the modified basic matrix to obtain an extended matrix of the LDPC code;

When ^, the extension matrix is used as the coding matrix of the LDPC code; when ^, the X^ ^xz/ G_" column on the right side of the extension matrix is intercepted, and the truncated matrix is used as the coding matrix of the LDPC code; or

When = ^3⁄4 is a positive integer, the unified base matrix is corrected to obtain a modified base \-R _a base matrix;

When ≥ _m , according to the expansion factor Z, the modified base matrix is extended by the unit matrix of zxz to obtain the coding matrix of the LDPC code; when <? _m , the +^ column to the right of the modified basic matrix is intercepted, According to the spreading factor Z, the matrix obtained by the truncation is expanded by using the unit matrix of ZXZ to obtain an encoding matrix of the LDPC code.

The information packet bit whose length is K is received, and the information bits to be encoded are determined as follows: The information packet bit of length K is directly used as the information bit to be encoded.

The coded mother code codewords are rearranged, wherein the mother code codeword bits before the rearrangement are Λ, -, ^, and the rearranged mother code codeword bits are Β ₀ Α ··, Β _Κ , —, The rearrangement formula is: · , where,

Otherwise nPad = [K'/z^zK' , PV represents the codeword rearrangement vector, its element set is {" /z,, " /z, + l, " /z, + 2, ~, " + 1} , PV(x) represents an element of index x in the PV vector, and X is a non-negative integer, where "^ represents an integer up to X;

When ≥ ^, the rearranged mother code codeword bits, A, · · ·, Β _κ , - are taken from the smallest to the largest in order to take _C C J J bits as the code word bits to be transmitted;

When ^ < R _m , the codeword bits of the mother code will be directly rearranged, Α,···, , _+Β3⁄4ΧΖ _^ as the codeword bits to be transmitted.

An apparatus for encoding an LDPC code, comprising: a coding unit, the apparatus further comprising: a comparison unit, a packet bit determining unit, an encoding matrix determining unit, an information bit determining unit, and a codeword processing unit;

a comparison unit, configured to compare a size of a mother code rate and a transmission code rate supported by the unified base matrix; a packet bit determining unit, configured to determine a length of the input information packet bit according to the comparison result of the comparison unit ;';

An encoding matrix determining unit, configured to determine an encoding matrix of the LDPC code according to the unified basic matrix and the spreading factor;

An information bit determining unit, configured to receive an information packet bit having a length of Κ, and determine an information bit to be encoded, and transmit the information bit to the coding unit;

And a codeword processing unit, configured to process the mother codeword obtained by performing LDPC encoding on the coding unit, to obtain a codeword bit to be transmitted.

The apparatus further includes: a setting unit configured to set a unified basic matrix of the LDPC code and an extension factor.

The device also includes an input unit for receiving a current transmission code rate;

≥W _m , the length K of the information packet bit is determined as (n _b -m _b )xz; when the comparison result of the comparison unit is < R _m , the length of the information packet bit is determined to be K' =

R), or K' = funcim^zlil-R^-m^z , or = func(func(m _b xz /(lR^xR ; where ^ is the transmission code rate, _Rm is the mother code rate, _mb , respectively, a unified basis The number of rows and columns of the matrix, z is the expansion factor, and /ifflc c) is an integer that takes an integer from X, an integer down, or a rounded integer.

The coding matrix determining unit is specifically configured to modify the unified basic matrix, and according to the spreading factor Z, use the unit matrix of Z X Z to extend the modified basic matrix to obtain

The coding matrix of the LDPC code. The information packet length of the received length K is directly used as the information bit to be encoded; when the comparison result of the comparison unit is ? _tt <? _m , the padding bit is added to the information packet bit of length K to form a length. The information bits to be encoded.

Row, where the mother code codeword bits before the rearrangement are, ^,...,^^, the rearranged mother code codeword bits are B ₀ A, , B _nbXz ― , and the rearrangement formula is:

^! S PV denotes a codeword rearrangement vector whose element set is ^+1, - +2,..., - 1} , PV(x) denotes an element of index X in the PV vector, and X is a non-negative integer;

When the comparison result of the comparison unit is R ≥ R _m , the rearranged mother code code word bit ^ is used. , ..., ^— ^ to take the index from small to large in order to take _1⁄2 _c ( / ? J bits as the codeword bits to be transmitted;

When the comparison result of the comparing unit <? _M, the added the k _b xz_K, mother code codeword bits after rearrangement ^ padding bits. , ..., ^— The system bits in i are deleted, and the remaining mother codeword bits are used as codeword bits to be transmitted.

The coding matrix determining unit is specifically configured to modify the unified basic matrix, and expand the modified basic matrix by using the unit matrix of z X z according to the spreading factor z to obtain the LDPC. The spreading matrix of the code; when the comparison result of the comparison unit is ≥? _m , the extended matrix is used as the coding matrix of the LDPC code; when the comparison result of the comparison unit is <? _m , the _1⁄2 3⁄4χζ/(1 of the right side of the extended matrix is intercepted - 》 column, the truncated matrix is used as the coding matrix of the LDPC code; or when k _b ' is a positive integer, the unified basic matrix is corrected to obtain the modified basic matrix;

\-R _a When the comparison result of the comparison unit is ≥, according to the expansion factor Z, the modified base matrix is expanded by the unit matrix of ZXZ to obtain the coding matrix of the LDPC code; when the comparison result of the comparison unit is <? _m Then, the +^ column on the right side of the modified basic matrix is intercepted, and according to the spreading factor z, the matrix obtained by the truncation is expanded by the unit matrix of ζ ζ , to obtain the coding matrix of the LDPC code.

The information bit determining unit is specifically configured to directly use the received information packet bit of length K as the information bit to be encoded.

The codeword processing unit is specifically configured to rearrange the mother codewords obtained by the coding unit, wherein the mother codeword bits before the rearrangement are 4, 4, . . . , 4 ₊ i, the rearranged mother Code code word bit is

Β ₀ , Β ..., Β _κ , _+η ^ , rearrangement formula is:

" A _k k <Κ' ^ , B = . , where ηΡαά =

, PV represents the codeword rearrangement vector, its element set is {" /z,, " /z, + l, " /ζ, + 2 ··, " + 1} , PV (x) represents the index in the PV vector For an element of x, X is a non-negative integer;

When the comparison result of the comparison unit is R ≥ R _m , the rearranged mother code code word bits ^ ..., ^ + ^ - i are taken from the smallest to the largest in order to take _1⁄2 _C ( / ? J bits, as Transmitted codeword bits;

When the comparison result of the comparison unit is <? _m , the mother code codeword bit ^ will be directly rearranged. , ^..., ^ ^ as the codeword bits to be transmitted. The LDPC code encoding method and apparatus provided by the present invention determine a basic matrix and a spreading factor, and perform processing of determining an input information packet length or a coding matrix according to the received transmission code rate to obtain a codeword that satisfies the foregoing transmission code rate. The variability of the code rate is realized on the basis of the optimal performance, the complexity of the hardware design for realizing different code rates is reduced, and the implementation cost is low. DRAWINGS

1 is a structural block diagram of a digital communication system;

2 is a schematic flowchart of an implementation of an LDPC code encoding method according to the present invention; FIG. 3 is a schematic flowchart of a first embodiment of an LDPC code encoding method according to the present invention; FIG. 5 is a schematic structural diagram of an apparatus for encoding an LDPC code according to the present invention. Detailed ways

The basic idea of the present invention is: comparing the mother code rate and the transmission code rate supported by the unified basic matrix, determining the length K of the input information packet bit according to the comparison result, and determining the coding of the LDPC code according to the unified basic matrix and the spreading factor a matrix; determining, by the received information packet bit of length K, the information bits to be encoded; performing LDPC encoding on the information bits to be encoded by using the coding matrix, and processing the encoded codeword bits of the encoded code to obtain The transmitted codeword bits.

The present invention will be further described in detail below with reference to the accompanying drawings.

FIG. 2 shows a flow of a method for encoding an LDPC code provided by the present invention. As shown in FIG. 2, the method includes the following steps:

S201, determining a unified basic matrix ^ and a spreading factor z of a fixed-size LDPC code;

In this step, the system can determine a unified base moment of size m _fi x according to actual needs. Array 1 and maximum spreading factor _2, where ^ is a positive integer, let = - ^, then the mother-coded rate supported by e is κ = κ; and the spreading factor is determined according to the length of the encoding code in the transmission of the digital communication system. Fixed constants, in general, to facilitate hardware implementation, the spreading factor is generally a power of two or a positive integer multiple of a power of two.

S202, receiving a current transmission code rate R _tx , comparing a size of R _t5 ^oR _m , and determining a length K of the input information packet bit according to the comparison result;

In this step, when ? _tt ≥ ? _m , the length of the information packet = ^Xz = (-)Xz; at that time, the parity bits generated by the encoding are transmitted, so the information packet length is Κ, and the parity bit can be The length m _fi xz and the transmission code rate R _{tx are} determined, so the relationship of K m _fi xz and R _tx can be, but is not limited to, expressed by the following formula: ^― ^ = R^ , K'+m _b xz = func(^^) ,

K'+m _b xz li? _tt or = f _unc (^^ ; Therefore, K, can be, but is not limited to, the following formula:

K' = func(m _b xzxR _a /(l- R _a )) , or K' = func(m _b xz /(I— m _b xz , or K' = func(func(m _b Xz/(lR xR; where /IOTC(JC) may represent an integer that takes an integer from x, an integer down, or a rounded integer. S203, determining an encoding matrix _{Hfe of the} LDPC code according to the spreading factor z and ; First, the above correction algorithm and the z pair are used to correct, and the base matrix A of the modified one is obtained, and the LDPC matrix H ^{p of} X z) XX z) is obtained according to the unit matrix of z and ζ χ , , which is both in the present invention. Hr ^p can be used as the coding matrix H _{fe of the} LDPC code, or can be intercepted according to the comparison result of 1^ and R _m to obtain the coding matrix H _{bz of the} LDPC code.

S204, determining, by the received information packet bit of length K, the information bit to be encoded; specifically, processing the information packet bit according to the size of the coding matrix and the comparison result of R _t oR _m to obtain information to be encoded Bit. S205: Perform LDPC encoding on the information bits to be encoded by the coding matrix H _fe to obtain a mother code codeword bit, and process the mother code codeword bits to obtain a codeword bit to be transmitted.

Specifically, the length of the codeword bit to be transmitted may be determined according to the transmission code rate, and the codeword bits to be transmitted are rearranged, shortened, or punctured according to the determined length of the codeword bit to be transmitted. . As shown in FIG. 3, the method includes the following steps:

S301-S302, determining a unified base matrix of a fixed-size LDPC code and a maximum spreading factor z _max , and determining the expansion factor Z as a fixed constant according to the required code length, and then executing S305;

In this step, the system can determine a unified base matrix of size m _fi x and a maximum expansion factor of ₂ according to actual needs, where ^, and _Zmax are positive integers, and = - ^, then the mother code code supported by e is obtained. The rate is κ - ^{m ln} _b = κ in _b . And the spreading factor z is determined to be a fixed constant according to the code length required for transmission in the digital communication system. Generally, in order to facilitate hardware implementation, the spreading factor z is generally 2 A power of a power or a positive integer multiple of a power of 2.

S303-S304, receiving the current transmission code rate R _tx , comparing the magnitudes of R _tx and R _m , and determining the length K of the input information packet bit according to the comparison result;

In this step, when ≥ ^, the length of the information packet bit = ^Xz = ( - )Xz ; when < ? _m , the parity bit of the generated code is transmitted, so the length of the information packet bit Κ, can be determined by the check bit length ^ Χζ and the transmission code rate R _tx , so K, m _fi Xz and

The relationship of R _tx can be, but is not limited to, expressed by the following formula: ^ - ^ = R _a , or

K'+m _b xz

K'+m _b xz = func(^^), or ^ = / _C (^3⁄4; Therefore, K, can be, but is not limited to, the following formula: K' = func(m _b xzxR _a /(l- R _a ) ), or K' = func(m _b z R^-m^z , or K' = func(func(m _b Xz/(lR )xR ; where / IOTC(JC) can represent an integer that takes an integer up to x, an integer down, or a rounded integer.

S305, correcting the unified basic matrix, obtaining the coding matrix H _{bz of the} LDPC code by using z extension, and then performing S308; in this step, first correcting the modified algorithm and the Z pair to obtain the corrected base matrix A, Obtaining an LDPC matrix of (m _b xz)x (n _b xz) according to the unit matrix of z and ζ χ , and using the matrix as the coding matrix of the LDPC code

H _bz .

S306-S307, determining, by the received information packet bit of length K, the information bits to be encoded;

Specifically, when ? _tt ≥ ? _m , the information packet bit of the received length K is directly used as the information bit to be encoded; when ? _tt < ? _m , padding is added to the information packet bit of length K. Bits, forming information bits to be encoded having a length of ^xz, wherein the method of adding padding bits may be before, after or between the information packet bits, and the length of the padding bits is k _b Xz-K, and, in addition, The processing is convenient, and the filled y^xz- bits are preferably all 0 bits, or may be all 1 bit.

_{S308-S309, χ (χ ζ} ) encoding matrix ^ ₂ treats information bits encoded by _¾χ ζ) LDPC encoding, to give mother code word bits of the mother code word bits, to give the codeword bits to be transmitted ;

Specifically, LDPC coding is performed by using the coding matrix H _fe to obtain a mother code codeword bit of length N = xz, wherein the first ^xz bits of the mother code codeword bits are system bits, and the last m _fi xz bits are checksums. Bit.

In order to improve the reliability of the codeword transmission, the codewords of the obtained mother code codeword bits may be first rearranged, and the codeword bits of the mother code before the rearrangement are assumed to be ^, ..., ^^, after the codewords are rearranged. Mother code The word bit is S. , A, · · ·, , then the codeword rearrangement is represented by the following formula:

Where PV represents a codeword rearrangement vector, and PV includes a different element whose element set is {n _b -m _b , n _b -m _b + n _b - m _b +2, ···, - 1} , And each element is different from each other, PV(x) means

An element of the PV vector indexed as X, where X is a non-negative integer.

When ≥ _{m m} , the puncturing operation is performed, that is, the mother code code word bits obtained after rearranging

^. , ..., ^— ^ According to the index from the smallest to the largest, take the _C ( / ? _tt ) bits as the codeword bits to be transmitted, where /iffl _C (X) can represent the integer up and down for X. Take an integer or round to take an integer. You can select the appropriate operation for / i« _C (x) according to the actual application.

When <? _m , the shortening operation is performed, and the y^xz- padding bits added in S307 are rearranged from the mother code codeword bits ^. , ..., ^ - i is deleted from the system bits (for the purpose of the deletion operation here, therefore, the padding bits are preferably all 0s or all 1 bits in S307), and the remaining mother code codeword bits are used as the code to be transmitted. Word bits.

The first embodiment of the encoding method of the LDPC code is specifically described below with reference to an example: Example 1:

1. Determine a unified basic matrix of 4x24 LDPC codes and a maximum spreading factor z max , 'where z, medium I z max = 96, ^{nb is} as follows ¹ ':

1 2S 55 -t 47 4 -1 91 84 S 15 52 .53 5 0 3⁄4 20 4 77 SO δ -ί -1

-1 6 -ί 3<5 4Q 47 12 79 47 -I 41 21 Π 71 72 0 44 49 0 Cs 0 Q -1

51 8:1 S3 4 67 -1 21 -1 31 24 91 61 Si § 6 7S SO SS 67 15 -! -1 0 0

-1 50 15 -1 13 16 Π 20 53 90 29 92 57 & S4 ©2 11 66 SO -I -IQ can be found ^ =4, n _b =2 , k _b =n _b -m _b =24-4 = 20 , the supported mother code rate is ^ = 20/24 = 5/6 = 0.833. In addition, in order to facilitate the realization and the actual required code length, the expansion factor z=3x2 ⁵ =96 is determined.

2. Receive the current transmission code rate R _tx =0.78 and compare the sizes of R _tx and R _m due to 0.78<0.833, so it can be obtained by ' = cez7(^xz/(1- ^xz, Κ, 1362, where ce/(x) means taking an integer up to X.

3. Correct the unified basic matrix by rounding the scale (scale+floor) algorithm and the expansion factor z.

■ γ τ modified

Get the corrected base matrix as follows:

25 55 -I 47 4 -1 91 S4 S 52 S2 33 5 & 36 2:3⁄4 4 77 SO S -! -1

6 -I 36 40 4? 12 7 & 47 -1 41 21 Ώ 71 U 72 0 4 49 (3 0 9 -1

S1 S3 4 67 -1 21 Λ 31 24 91 61 SI 9 M 7S ω' S2 67 15 -I -1 0 0

-1 50 15 -1 36 13 IS 11 20 53 90 29 92 57 MM 92 11 66 m -1 -IQ Obtain an (4x96)x(24x96) LDPC matrix Η ^ρ according to the unit matrix expansion with z=96 and zx z And the matrix Η ^{ρ is taken} as the coding matrix H _{bz of the} LDPC code.

4. Receive the information packet bit of length K, because 0.78O.833, ie _tt < ? _m , fill the length of ^xz- = 558 all 0 bits into the front of the information packet bit of length K. , the information bits to be encoded of length k _b xz = 1920 are obtained.

5. LDPC encoding the information bits to be encoded with a length of xz = 1920 by H _fe of (4 X 96) X (24 x 96) to obtain a mother code codeword bit of length N = n _b x _z = 2304 , where the first 1920 bits are system bits and the last 384 bits are parity bits.

The codeword rearrangement is performed on the mother code codeword bits, and it is assumed that the mother code codeword bits before the rearrangement are _A0 , A-, _A2im , and the mother code codeword bits after the codeword rearrangement are Α, ^··· ^^, then the codeword rearrangement is represented by the following formula:

Where PV represents the codeword rearrangement vector, PV = {20, 21, 22, 23}. Also, since 0.78 < 0.833, that is, <i? _m , the mother code codeword bits obtained by rearranging 558 all 0 bits in front of the information packet bits are rearranged. , ₁ ,..., . The system bits in ₃ are deleted, and the remaining mother code code words are S ₅₅₈ , ₉ , ···, . ₃ as the codeword bit to be transmitted.

Example 2: 1. Determine a unified base matrix of 4x24 LDPC codes and a maximum spreading factor z max , 'its z, medium I z max = 96, ^{Mb is} as follows ¹ ':

1 25 55 -I 47 4 -1 91 U S 52 S2 33 5 & 36 2:3⁄4 4 77 SO S -! -1

-1 6 -I 36 40 4? 12 7 & 47 -1 41 21 O 71 14 72 0 44 49 (3 0 9 -1

51 SI S3 4 67 -1 21 -1 31 24 91 61 SI 9 M 7S SO g£ 67 15 -! -1 0 0

§S -1 15 -1 36 13 IS 11 20 53 90 29: θ2 57 MM 92 H 66 SO -1 -IQ can get _{fi =} 4 , n _b =24, k _b =n _b -m _b =24-4 = 20 , The supported mother code rate is ? _m = 20/24 = 5/6 = 0.833 , and for ease of implementation, the expansion factor ζ = 2 ⁶ = 64 is determined.

2. Receive the current transmission code rate R _tx =0. 875, compare the size of 1^ and 1^, since 0. 875>0.833, the information packet length = k _b xz = 1280.

3. Correct the unified base matrix by rounding the scale+floor algorithm and the expansion factor z to obtain the modified base matrix A as follows:

Q 16 36 -1 31 2 -1 6Θ 56 5 57 34 S 22 3 0 2 13 2 51 53 Θ -1 - 1

- 1 h -1 2 26 31 8 52 31 - 1 27 14 8 47 9 48 β 29 32 ø■ Θ & β -1

34 54 55 2 Η -1 13⁄4 - 1 20 16 6Θ 40 54 6 57 52 48 58 44 10 -1 - 1 :ø■ β

45 -1 33 10 -1 2 8 6 7 13 35 6β 19 61 38 2Θ 56 61 7 itU 53 - 1 -1 β According to the unit matrix of z=64 and zxz, an LDPC matrix of (4χ64)χ(24χ64) is obtained. Η ^ρ , and the matrix Η ^{ρ is} used as the coding matrix H _{bz of the} LDPC code.

4. The information packet bit of length K, =1280 is received. Since 0.875>0.833, the information packet bit of the length ^}K' = k _b xz = 1280 is directly used as the information bit to be encoded.

5. LDPC encoding the information bits to be encoded with a length of ' = xz = 1280 by H _bz of (4 X 64) X (24 x 64) to obtain a mother codeword bit of length N = xz = 1536, The first 1280 bits are system bits, and the last 256 bits are parity bits.

The codeword rearrangement is performed on the mother code codeword bits, and it is assumed that the mother code codeword bits before the rearrangement are ^,...,^^, and the mother code codeword bits after the codeword rearrangement are Α,Α,··· , Α ₅₃₅ , then the codeword rearrangement is represented by the following formula: A _k ^<1280

Otherwise where PV represents the codeword rearrangement vector, PV = {20, 21, 22, 23}. And because 0.875>0.833, which belongs to ≥? _m , the mother code code word bit S will be rearranged. According to the index, the _3⁄4or( / ) = 1462 bits are taken from the smallest to the largest, that is, 3⁄4, Α, . . . , _{61 is} used as the codeword bit to be transmitted, where ^r ( ^X ) represents an integer from X downward. . As shown in FIG. 4, the method includes the following steps:

S401-S402, determining a unified basic matrix of a fixed-size LDPC code and a maximum spreading factor z _max , and determining the expansion factor Z as a fixed constant according to the required code length; in this step, the system may determine one according to actual needs. a unified base matrix A of size m _fi X and a maximum expansion factor z _max , where ^, and z _max are positive integers,

Then, the mother code rate supported by e is ^{-m lr} =K ln _b . And the spreading factor z is determined to be a fixed constant according to the code length of the digital communication system in the transmission. Generally, in order to facilitate hardware implementation, the expansion factor z - is a power of 2 or a positive integer multiple of a power of 2.

S403, receiving the current transmission code rate R _tx , and then performing S404 and S405 respectively;

S404, comparing the size of 1^ and 1^, according to the comparison result, determining the length of the input information packet bit Κ ', and then executing S406;

In this step, when ≥ ^, the length of the information packet bit = ^xz = ( - ) xz; when < ? _m , the parity bits of the generated code are transmitted, so the length of the information packet bit Κ, can be determined by the check bit length ^ χζ and the transmission code rate R _tx , so K, m _fi xz and

K'+m _b xz

K'+m _b xz = func(^^). or ^ = / _C (^3⁄4; Therefore, K, can be, but is not limited to, the following formula Represents: K' = funcim^zxR^ /(I- R _a y) , or K' = func(m _b z R^-m^z , or K' = func(func(m _b Xz/(lR )xR ; / IOTC(JC) can represent an integer that takes an integer from x, an integer down, or a rounded integer.

,

S405, correcting the unified basic matrix, determining the coding matrix H _{fe of the} LDPC code, and then performing S407; in this step, first correcting the modified algorithm and the z pair to obtain the corrected base matrix A, according to and And the unit matrix of ζ χ 扩展 is extended to obtain the LDPC extension matrix Η ^{ρ of} (m _b xz)x(n _b xz); when ≥ _{m m} , the extension matrix Η ^{ρ is} used as the coding matrix H _{bz of the} LDPC code; ? _m , by intercepting the right f leg c (m _b xz / (l_RJ) 歹 ^, the size is

(?3⁄4χζ)χ_1⁄2 3⁄4χζ/(1- The coding matrix. Also K'+m _b xz = func(^^) , therefore,

1 - The size of the coding matrix of the LDPC code at this time is ( x _z )x( +^xz). In addition, when =^3⁄4 is a positive integer, after correcting the unified basic matrix ^∞73⁄4 TM into the lR _a row in this step, the corrected basis may also be obtained according to the comparison result of 1^ and R _m .

After the matrix is processed, it is expanded by the spreading factor z to obtain an encoding matrix H _{fe of the} LDPC code.

Specifically, when ? _tt ≥ ? _m , the modified base matrix is expanded, and the coding matrix is specifically the same as the method of obtaining the coding matrix by the above method, and will not be described again; when < ^, by intercepting ^δ right The k _b '+m _b column gives a matrix H _fi ^m ° ^di/M - of size ^ x( + ); and the expansion factor z, the extension gives ( _WFI X _Z )X(( + )XZ The coding matrix of the LDPC code. S404, ^ R _a <R i, K'+m _b xz = func(^^-) ,

1-^ And _k ' is a positive integer, so you can get _K ' _+M

, ie l_i? _tt lR^

(k _b ' +m _b )xz = func^^- = K'+m _b xz, so at this time, the obtained coding matrix l_i? of the LDPC code,

The size is also (m _b xz x(K, +m _b xz .

S406. Receive an information packet bit of length K, and directly use the information packet bit as the information bit to be encoded.

S407-S408, performing LDPC encoding on the information bits to be encoded by the coding matrix _Hfe , obtaining a mother code codeword bit, and processing the mother codewordword bits to obtain a codeword bit to be transmitted.

Specifically, when ≥ _m , the information bit to be encoded is LDPC-encoded by the coding matrix matrix _Hbz of the LDPC code of size X z) XX z) to obtain a mother code codeword bit of length N = + Xz = Xz Wherein the first bit of the mother code codeword bit is a systematic bit, and the last m _fi xz bits are check bits, wherein the length of the information bit to be encoded is

K' = k _b Xz = (n _b -m _b )xz.

In order to improve the reliability of the codeword transmission, the codeword rearrangement of the obtained mother code codeword bits is first performed, and the mother code codeword bits before the rearrangement are ^, ..., ^^+^^, the codeword weight The rear coded codeword bits are ^, ...,^, then the codeword rearrangement is represented by the following formula:

Where "Λ3⁄4 = "^/ _ζ ]*ζ- , PV represents the codeword rearrangement vector, PV contains ^ different elements, and its element set is {" /z], " /z] + l, " /z ] + 2, —, " /ζ] + ^_1} , and each element is different from each other. PV(x) represents an element of the PV vector indexed by X, where X is a non-negative integer.

When ? _tt ≥ ? _m , the puncturing operation is performed, that is, the mother code code word bits ^, ^..., ^^^ obtained after the rearrangement are taken as / _C ( / ? _tt ) bits according to the index from small to large, As to be transmitted The codeword bits, ie, ^, ..., ^, ^ - i are the codeword bits to be transmitted. Xx) can represent an integer that takes an integer from X, an integer down, or a rounded integer. Specifically, an appropriate operation can be selected for /IfflC(X) according to the actual application.

When <? _m , since the interception operation has been performed in the process of obtaining the coding matrix ^ in S405, the mother code codeword bits, ^..., ^^ obtained after the rearrangement can be used as the codeword bits to be transmitted.

The second embodiment of the encoding method of the LDPC code is specifically described below with reference to an example: Example 3:

1. Determine a unified basic matrix of 6x24 LDPC codes and a maximum spreading factor z, 'its, z = 96, ^Mb are as follows, :

6: 38 3 93 -1. -ί -1 30 70 -1 86 -1 37 38 4 11 -1 46 4S 0 -I -I -1 -1 62 94 19 S4 -1 92 7S -i :15 -1 -I 92 -1 45 24 32 30 -1 -1 0 0 -1 -I -1

71 -1 55 -I 12 66 45 ?9 -I 78 -1 -1 10 -1 22 5.5 ?0 B2 -i - 0 0 -1 -1

3S 61. -1 66 9 ?3 47 64 -I 39 6J 43 -1 -1 -I -1 95 32 0 - -1 0 0 -1

-! -1 -1 -I 32 2 55 SO 95 2 6 51 24 90 44 2C -ί -I -1 - -1 -I 0 0

-i 63 3i SS 20 -t -1 -i 6 40 S6 !6 7i 53 -I -! 27 2.6 4S - -i -1 -1 0 Can you get? _{¾ = 6, n b = 24} , k b = n b -m b = 24-6 = 18, Mb support mother code rate of _{m = (-? = ^ /} ¾ = 18/24 = 3/4 = 0.75, and in order to facilitate the implementation and the actual required encoding code length, determine the expansion factor _z = 3x2 ⁵ = 96.

2. Receive the current transmission code rate R _tx =0.9 and compare the sizes of R _tx and R _m . Since 0.9>0.75, the information packet length = X _z = 1728.

■ ττ

Get the corrected base matrix A as follows:

6 s 93 -1. -I -1 3 70 -1 -1 38 4 11 -1 46 4S 0 -I -i -1

62 94 19 S4 -1 92 ; -I 15 -1 -l 92 -1 45 24 2 30 -1 -1 0 0 -1 -1

71 -1 55 - i 12 66 45 ?9 -I -I -1 10 -1 ': 5 70 B2 -1 -1 0 0 -1

61. -1 66 ?3 47 64 -I 39 61 43 -1 -1 -1 -1 o -I -1 0 0 -1

- 1 -1 -1 2 SO 6 51 90 44 20 -i -1 -1 -1 — i -1 0 0

63 3i 8S 20 -1 6 40 S6 16 ?i 53 -1 2.6 S - -1 -1 0 Ττ modified

According to the unit matrix expansion of z=96 and zxz, the LDPC matrix Η ^{ρ of} (6x96)x(24x96) is obtained, and 0.9>0.75, then the matrix Η ^{ρ is} used as the coding matrix H _{bz of the} LDPC code.

4. Receive the information packet bit of length K,=1728, and the length is =

= 1728 The information packet bit is directly used as the information bit to be encoded.

5. LDPC encoding the information bits to be encoded with (6x96)x(24x96) H _fe † length K' = k _b xz = m^, and obtain a length of N = +m _fi x _z = xz = 2304 The mother code code word bits, wherein the first 1728 bits are system bits, and the last 576 bits are school risk bits.

The codewords are rearranged for the mother codeword bits, assuming that the mother codeword bits before the rearrangement are 4, -, ₃ ο3 'the codeword bits after the codeword rearrangement are 3⁄4, ..., . ₃ , then the codeword rearrangement is represented by the following formula:

Where "Λ3⁄4 ="'/ζ,*ζ-' = 0 , PV represents the codeword rearrangement vector, PV={18,19,20,21,22,23}„ and since 0.9>0.75, belongs to? _tt ≥ ? _m , will be rearranged to get the mother code codeword bits 3⁄4, A, · · ·, Β ₂₃₀₃ according to the index from small to large, take cd K '/R = l 920 bits, that is, ^, ..., ₁₉ as The codeword bits to be transmitted, where _ce /(x) represents an integer up to X.

Example 4:

6 38 ji -1 -1 -1 30 70 -1 S6 -1 37 4 11 -1 46 48 0 - i -1

62 94 19 84 -1. 92 78 -1 15 -1 -Ϊ 92 - 1 45 24 30 -1 -I 0 0 -1 -i "; -1 55 : 12 66 45 79 -I 7S -I -1 io -: 1 5 70 S2 -I -] 0 ■Q - 1

38 1 -1 9 47 64 -I 39 61 43 -i - ί -i 95 32 0 -1 -1 G 0

-i - 1. -1 - 1 80 95 6 51 24 90 44 20 -1 - 1 - i -1 0

_1 63 31 2 -i -i -1 6 40 56 i 6 S3 -1 ― i 27 26 4 i -1 -i -1 It can be found that m _b =6, = 24, a 24-6 = 18 supported mother code rate _{_{R, ln b = ^ / ¾}} = 18/24 = 3/4 = 0.75, and the other to achieve the actual needs in order to facilitate the provision The code length is long, and the expansion factor _z = 2 ⁶ = 64 is determined.

2. Receive the current transmission code rate R _tx =0.6, compare the size of 1^ and 1^, since 0.6<0.75, you can find K by = ce/(^xz/(l- D)- ?3⁄4xz, Is 576, where ce/(;c) represents an integer up to X.

3. Correct the unified base matrix ^ by rounding the scale+floor algorithm and the expansion factor z to obtain the modified base matrix A as follows:

25 2 62 -1 -1 -1 20 46 -1 57 -1 2 25 2 7 -1 30 32 β -1 -1 -1 -1

ΗΛ 62 12 56 -1 61 52 -1 10 -1 -1 61 -1 30 16 21 20 -1 -1 β β -1 -1 -1 i -1 36 -1 8 3Θ 52 -1 52 -1 -1 6 -1 14 36 46 Sh -1 -1 β β -1 -1

25 -1 44 6 48 31 42 -1 26 4β 28 -1 -1 -1 -1 63 21 β -1 -1 β β -1

-1 -1 -1 -1 21 34 36 53 63 14 h 3U 16 6β 2 13 -1 -1 -1 -1 -1 -1 β β

-1 2β 58 13 -1 -1 -1 h 26 37 10 47 35 -1 -1 18 17 32 -1 -1 -1 -1 β According to the unit matrix expansion with z=64 and zx z (6χ64)χ The LDPC matrix (7 of (24χ64), and since 0.6<0.75, intercepts the ceil (m _b xz/(lR ) = 960 columns to the right of Η ^{ρ to} obtain the coding matrix H _{bz of} size (6x64) x960. In addition, this step It can also be processed as follows: Get the corrected base matrix, and since 0.6O.75, k _b ' =^^ = 9 , intercept the right +^=9 + 6 = 15 columns, and get the size of 6x(9 + 6) The matrix d - ; and z = 64 and the unit matrix of zz are extended to obtain the coding matrix ^ of the (6x64) x (15x64) LDPC code.

4. The information packet bit of length K is received, and the information packet bit of length 576 is directly used as the information bit to be encoded.

5. Perform (6x64)x(15x64)^ ₂ pairs of information bits to be encoded with length = 576

LDPC coding, obtaining a mother code codeword bit of length N = + x _z = xz = 960, where the former

576 bits are system bits and the last 384 bits are parity bits.

Perform codeword rearrangement on the mother codeword word bits, assuming that the mother code codeword bits before the rearrangement are 4, -, ₅ 9 'the codeword rearranged mother code codeword bits are Α, Α, ··· , ₅₉ , the codeword rearrangement is represented by the following formula:

Otherwise

Where ",3⁄4/ ="'/ζ,*ζ-' = 0 , PV represents the codeword rearrangement vector, PV={9,10,11, 12,13, 14}. Also due to 0.6<0.75, ie R <R _m , since the interception operation has been performed in the process of obtaining the coding matrix in 3, the mother code code word bits obtained after the rearrangement, Α, . . . , _{59 can be} used as the code word bits to be transmitted.

FIG. 5 shows the structure of an apparatus for encoding an LDPC code provided by the present invention. As shown in FIG. 5, the apparatus includes: an encoding unit 10, a comparing unit 20, a packet bit determining unit 30, an encoding matrix determining unit 40, and information bits. a determining unit 50 and a codeword processing unit 60; wherein the comparing unit 20 is configured to compare the size of the mother code rate and the transmission code rate supported by the unified base matrix; the grouping bit determining unit 30 is configured to determine according to the comparison result of the comparing unit 20 The length of the input information packet bit Κ, the encoding matrix determining unit 40 is configured to determine an encoding matrix of the LDPC code according to the unified basic matrix and the spreading factor; the information bit determining unit 50 is configured to receive the input information bit, and determine according to the packet bit determining unit 30. The information packet bit length is obtained, the information packet bit of length K is obtained, and the information bits to be encoded are determined and transmitted to the coding unit 10; the codeword processing unit 60 is used to perform LDPC coding on the coding unit 10 The codeword word is processed to obtain the codeword bits to be transmitted.

Further, the apparatus further includes a setting unit 70 configured to set a unified base matrix and an expansion factor of the LDPC code.

Further, the device further includes an input unit 80, configured to receive a current transmission code rate, and the packet bit determining unit 30 is specifically configured to determine a length K of the information packet bit when the comparison result of the comparison unit 20 is R ≥ {n _b -m _b )xz; when the comparison result of the comparison unit 20 is R <R _m , the length of the information packet bit is determined to be Κ, =

R ) , or K' = func(m _b Xz / lR^-m^z , ^K' = func(func(m _b z /(I- R^ R^); where R is the transmission code rate, Mother code rate, _mb , respectively, the number of rows and columns of the unified base matrix, z For the expansion factor, /ifflc c) is an integer that takes an integer up to x, an integer down or rounds to an integer. Further, the coding matrix determining unit 40 is specifically configured to modify the unified basic matrix, and according to the spreading factor z, extend the modified basic matrix by using the unit matrix of zxz to obtain an encoding matrix of the LDPC code.

Correspondingly, the information bit determining unit 50 is specifically configured to: when the comparison result of the comparing unit 20 is ≥, the information packet bit of the received length K is directly used as the information bit to be encoded; when the comparison result of the comparing unit 20 is When < _m , a padding bit is added to the information packet bit of length K to form an information bit to be encoded of length ^xz, wherein the padding bit is all 0 bits or all 1 bit.

Correspondingly, the codeword processing unit 60 is specifically configured to rearrange the mother codewords obtained by the encoding unit 10, wherein the mother codeword bits before the rearrangement are, ^, ..., ^^, rearranged mother The codeword bits are BoU^ and the rearrangement formula is:

Rearrange the vector, its element set is ^+1, - +2,..., - 1} , PV(x) represents the element with the index X in the PV vector, X is a non-negative integer; when the comparison result of the comparison unit 20 When R _a ≥R _m , the rearranged mother code codeword bits ^, ^..., ^^ are taken from the smallest to the largest in order to take 1⁄2 _c (/D bits as the codeword bits to be transmitted; The comparison result of unit 20 is

When R _< R _m , the added y^xz- padding bits are added from the rearranged mother code codeword bits. , ^···, ^^ is deleted from the system bits, and the remaining mother code codeword bits are used as codeword bits to be transmitted.

Further, the coding matrix determining unit 40 is specifically configured to modify the unified basic matrix, and according to the spreading factor Z, the modified basic matrix is extended by using the unit matrix of zxz to obtain an extended matrix of the LDPC code; when comparing the comparison unit 20 When the result is ≥ _m , the extended matrix is used as the coding matrix of the LDPC code; when the comparison result of the comparison unit is <? _m , the interception is performed. Extend the X^ ^xz/(1 - 》 column on the right side of the matrix, and use the truncated matrix as the coding matrix of the LDPC code; or when the =^3⁄4 is a positive integer, correct the unified base matrix and get the correction.

\ Base matrix after -R _a; when the comparison result of the comparing unit is ≥ 20, according to the spreading factor Z, using z X z a unit matrix obtained by correcting the base matrix is extended to give the coding matrix of the LDPC code; when comparing When the comparison result of the unit 20 is R _< R _m , the +^ column on the right side of the corrected basic matrix is intercepted, and the truncated matrix is expanded by the unit matrix of zxz according to the spreading factor z to obtain an encoding matrix of the LDPC code.

Correspondingly, the information bit determining unit 50 is specifically configured to directly use the received information packet bit of length K as the information bit to be encoded.

Correspondingly, the codeword processing unit 60 is specifically configured to rearrange the mother codewords obtained by the encoding unit 10, wherein the mother codeword bits before the rearrangement are ^), ^, ..., ^, rearranged mother The code word bits are Β ₀ Α,···,Β _Κ ,— , and the rearrangement formula is:

ΡΡαά =

, PV represents the codeword rearrangement vector, its element set is {" /z,, " /z, + l, " /z, + 2, ~, " + 1} , PV (x) represents the index in the PV vector X is a non-negative integer for the element of x; when the comparison result of the comparison unit 20 is ≥ ? _m , the rearranged mother code codeword bits β ₀ Ί··, β _κ , - are indexed from small to large in order Taking f K, IRJ bits as the codeword bits to be transmitted; when the comparison result of the comparison unit 20 is < ^, the mother code codeword bits will be directly rearranged. A, · · · ·, Β _κ , as the codeword bits to be transmitted.

It should be noted that the encoding and decoding are mutually corresponding. Therefore, the encoding method and apparatus for the LDPC code provided by the present invention have corresponding inverse operations on the decoding method and the decoding device, and are not described again.

The above is only the preferred embodiment of the present invention and is not intended to limit the scope of the present invention.

Claims

Claim

A method for encoding a low density parity check LDPC code, the method comprising the steps of:

Comparing the mother code rate and the transmission code rate supported by the unified base matrix, and determining the length K of the input information packet bit according to the comparison result;

Determining information bits to be encoded by receiving information packet bits of length Κ;

The information bits to be encoded are LDPC-encoded by the coding matrix, and the encoded mother code codeword bits are processed to obtain codeword bits to be transmitted.

2. The method according to claim 1, wherein the method further comprises: setting a unified basic matrix and an expansion factor of the LDPC code.

The method according to claim 1 or 2, wherein before the comparing the code rate of the mother code and the size of the transmission code rate supported by the unified base matrix, the method further includes: receiving the current transmission code rate;

When ≥ _m , the length K of the information packet is determined as - m _b )xz; when <, the length of the information packet bit is determined to be K ' = funcimb X zx ^ /(\- R ) , or K ' = func (m _b xz /(1 - )) - m _fi xz or = func(func(m _b xz /(l - )) x );

Wherein, for the transmission code rate, R _m is the mother code rate, m _b , respectively is the number of rows and columns of the unified base matrix, Z is the expansion factor, /iffl _C (X) is the integer from X up, down Take an integer or round to take one of the integers.

The method according to claim 1 or 2, wherein the determining an encoding matrix of the LDPC code according to the unified basic matrix and the spreading factor is:

Correcting the unified base matrix, using the unit matrix pair of zxz according to the expansion factor z The modified basic matrix is expanded to obtain an encoding matrix of the LDPC code.

The method according to claim 4, wherein the information packet bits of the received length K are determined to be:

When ≥, the received information packet length is 直接, and the information packet bit is directly used as the information bit to be encoded; when <? _m , the padding bit is added to the information packet bit of length K to form a length of ^Χζ Information bits to be encoded.

The method according to claim 5, wherein the processing of the encoded mother code codeword bits is performed as follows:

The coded mother code code words are rearranged, wherein the mother code code word bits before the rearrangement are A„A _x , -, A _nb ^ rearranged mother code code word bits are ^ ..., ^, rearrangement The formula is:

^l », PV represents the codeword rearrangement vector, its element set is ^+1, - +2,..., - 1} , PV(x) represents the element of the PV vector index X, and X is a non-negative integer;

When ≥ ^, put the rearranged mother code code word bits. , A, · · ·, according to the index from small to large, take the bits in order, as the codeword bits to be transmitted;

When < ^, the k _b Xz — K to be added, the padding bits are from the rearranged mother code codeword bits. The system bits in A, · · ·, are deleted, and the remaining mother code codeword bits are used as codeword bits to be transmitted.

The method according to claim 3, wherein the determining the coding matrix of the LDPC code is:

Correcting the unified basic matrix, according to the expansion factor z, using the unit matrix of zxz to extend the modified basic matrix to obtain an extended matrix of the LDPC code;

When ≥ _m , the extended matrix is used as the coding matrix of the LDPC code; when <? _m , the Χ^ ^χζ/(1_ 》 column on the right side of the extended matrix is intercepted, and the truncated matrix is taken as the LDPC code. Encoding matrix; or,

When ≥, according to the expansion factor Z, the modified base matrix is extended by the ZXZ unit matrix to obtain the coding matrix of the LDPC code; when <? _m , the +^ column on the right side of the modified basic matrix is intercepted, according to the extension The factor Z is extended by the unit matrix of ZXZ to obtain the coding matrix of the LDPC code.

The method according to claim 7, wherein the information packet bit of the received length K is determined to be: the information packet bit of length Κ is directly used as the information packet to be encoded. Information bits.

9. The method according to claim 8, wherein the processing of the encoded mother code codeword bits is performed as follows:

The coded mother code codewords are rearranged, wherein the mother code codeword bits before the rearrangement are Λ, -, ^, and the rearranged mother code codeword bits are Β ₀ Α ··, Β _Κ , —, The rearrangement formula is:

" A k < K' ^ , B = . , where

( ηΡαά =

, PV represents the codeword rearrangement vector, its element set is {" /z,, " /z, + l, " /z, + 2, ~, " + 1} , PV (x) represents the index in the PV vector For an element of x, X is a non-negative integer, where "^ means taking an integer up to X;

When ≥ ^, put the rearranged mother code code word bits. , A, · · · ·, Β _Κ , - take 0) bits in order from the _smallest to the largest, as the codeword bits to be transmitted;

When ^ < R _m , the codeword bits of the mother code will be directly rearranged. , Α,···, , _+Β3⁄4ΧΖ _^ is the codeword bit to be transmitted.

10. An encoding apparatus for an LDPC code, comprising: an encoding unit, wherein the loading The method further includes: a comparing unit, a grouping bit determining unit, an encoding matrix determining unit, an information bit determining unit, and a codeword processing unit; wherein

a comparison unit, configured to compare a size of a mother code rate and a transmission code rate supported by the unified base matrix; a packet bit determining unit, configured to determine a length of the input information packet bit according to a comparison result of the comparison unit;

An information bit determining unit, configured to receive the obtained information packet bit of length Κ, and determine the information bit to be encoded, and transmit the information bit to the coding unit;

The device according to claim 10, wherein the device further comprises: a setting unit, configured to set a unified basic matrix and an expansion factor of the LDPC code.

The device according to claim 10 or 11, wherein the device further comprises an input unit, configured to receive a current transmission code rate; and determine a length K of the information packet bit, which is (n _b - m _b ) Xz; When the comparison result of the comparison unit is < R _m , the length of the information packet bit is determined to be Κ, =

R ) , or K, =

, or ' = func(func(m _b X z /(l - R^ R^); where ^ is the transmission code rate, _Rm is the mother code rate, _mb , respectively, the number of rows and columns of the unified base matrix , z is the expansion factor, /ifflc c) is an integer that takes an integer up to X, an integer down, or a rounded integer.

The apparatus according to claim 10 or 11, wherein the encoding matrix determining unit is specifically configured to correct the unified basic matrix, and according to the expansion factor z, the corrected using the unit matrix pair of zxz The base matrix is extended to obtain an encoding matrix of the LDPC code.

The device according to claim 13, wherein the information bit determining unit is specifically configured to directly receive the information packet length of the received length K when the comparison result of the comparing unit is ≥? _m As information bits to be encoded; when the comparison result of the comparison unit is < W _m , padding bits are added to the information packet bits of length K to form information bits to be encoded having a length of k _b xz.

The apparatus according to claim 14, wherein the codeword processing unit is specifically configured to rearrange a mother code codeword obtained by coding the coding unit, wherein the mother code codeword bit before the rearrangement is 4), 4,···, ^— i , the rearranged mother code codeword bits are Βο, Β^-, Β^, and the rearrangement formula is:

When the comparison result of the comparison unit is R ≥ R _m , the rearranged mother code codeword bits are used.

Α,Α,···^^—^ It is necessary to take a bit from the smallest to the largest according to the index, as the codeword bits to be transmitted;

When the comparison result of the comparing unit is _<m, the added the k _b Xz_K, mother code codeword bits after rearrangement padding bits Α, ^ ··· ^^ -? I in the systematic bits deleted The remaining mother code codeword bits are used as codeword bits to be transmitted.

The apparatus according to claim 12, wherein the coding matrix determining unit is specifically configured to modify the unified basic matrix, and expand the modified basic matrix by using the unit matrix of zxz according to the spreading factor z. Obtaining an extension matrix of the LDPC code; when the comparison result of the comparison unit is ≥? _m , the extension matrix is used as the coding matrix of the LDPC code; when the comparison result of the comparison unit is <, intercepting the X^ ^XZ - ^R on the right side of the extension matrix -" 歹|J , the truncated matrix is used as the encoding matrix of the LDPC code; or ==3⁄4 is a positive integer

\-R _a

When the unified base matrix is modified, the modified basic matrix is obtained. When the comparison result of the comparison unit is ≥, according to the expansion factor Z, the modified base matrix is extended by the ZXZ unit matrix to obtain the coding of the LDPC code. Matrix; When the comparison result of the comparison unit is <, the +^ column on the right side of the modified base matrix is intercepted, and according to the expansion factor Z, the matrix obtained by the truncation is expanded by the unit matrix of zxz to obtain the coding matrix of the LDPC code.

The device according to claim 16, wherein the information bit determining unit is specifically configured to directly use the received information packet bit of length K as the information bit to be encoded.

The apparatus according to claim 17, wherein the codeword processing unit is specifically configured to rearrange the mother codewords obtained by the coding unit, wherein the mother codeword bits before the rearrangement are 4 , 4, _···· ,4, _+Β3⁄4χζ —, the rearranged mother code codeword bits are Β ₀ Α ··,Β _Κ , —, rearrangement formula is:

B _k · , where,

Otherwise

ΡΡαά =

The word rearrangement vector, whose element set is {" /z,, " /z, + l, " /z, + 2, ~, " + 1} , PV(x) represents the element indexed x in the PV vector , X is a non-negative integer;

^ ...,^+^— i takes the bits in order from the smallest to the largest, as the codeword bits to be transmitted;

When the comparison result of the comparison unit is <, the codeword bits of the mother code will be directly rearranged from

^ ,..., ^^ as the codeword bits to be transmitted.