KR101413783B1 - Apparatus and method for channel encoding and decoding in communication system using variable-length ldpc codes - Google Patents

Abstract

A method of encoding a structured LDPC code having a variable length, the method comprising: detecting first parity check matrix information stored in advance; checking a block length of a required LDPC code; The method comprising the steps of: determining a size to group information words of a matrix; constructing a partial matrix corresponding to an information word of a second parity check matrix from the first parity check matrix to correspond to the grouping size; Constructing a partial matrix corresponding to the parity of the check matrix and performing encoding based on the first parity check matrix or the newly generated second parity check matrix.

LDPC coding, variable block length, shortening, puncturing

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a channel encoding / decoding method and apparatus using a low-density parity-check code having a variable block length,

The present invention relates to a communication system applying a low-density parity-check (LDPC) code as an error correction code, and more particularly to a channel encoding / decoding method using an LDPC code having a variable block length, decoding apparatus and method.

In wireless communication systems, the performance of a link is significantly degraded due to various noise, fading phenomena, and inter-symbol interference (ISI) of a channel. Therefore, it is essential to develop overcoming techniques for noise, fading and ISI in order to realize high-speed digital communication systems requiring high data throughput and reliability, such as next generation mobile communication, digital broadcasting, and portable Internet. In recent years, error-correcting codes have been actively studied as methods for efficiently restoring information distortion and improving communication reliability.

The LDPC code first introduced by Gallager in the 1960s has long been forgotten due to its implementation complexity far surpassing that of the time. However, since the turbo codes found by Berrou, Glavieux, and Thitimajshima in 1993 show performance close to the channel capacity of Shannon, many interpretations on the performance and characteristics of turbo codes have been made, and it has been studied a lot about iterative decoding and graph based channel coding. As a result, the LDPC code is re-studied in the late 1990's and applied iterative decoding based on a sum-product algorithm on a Tanner graph (a special case of a factor graph) corresponding to the LDPC code, Has a performance close to Shannon's channel capacity.

The LDPC codes are typically represented using graphical representations, and many features can be analyzed through graph theory, algebra, and probability-based methods. Generally, a graph model of a channel code is useful not only for describing codes but also to correspond information of encoded bits to a vertex in a graph and to correspond each bit relation to edges in a graph , It is possible to derive a natural decryption algorithm because each vertex can be regarded as a communication network for exchanging messages determined by each line. For example, trellis-derived decoding algorithms, which can be regarded as a kind of graph, include the well-known Viterbi algorithm and BCJR (Bahl, Cocke, Jelinek and Raviv) algorithms.

The LDPC code can be expressed using a bipartite graph, which is generally defined as a parity-check matrix and is collectively referred to as a Tanner graph. The bipartite graph indicates that the vertices constituting the graph are divided into two different types. In the case of the LDPC code, a binary graph composed of a variable node and a vertex called a check node Is expressed. The variable node corresponds one-to-one to the encoded bit.

A graphical representation method of the LDPC code will be described with reference to FIGS. 1 and 2. FIG.

1 is an example of a parity check matrix H1 of the LDPC code including four rows and eight columns. Referring to FIG. 1, an LDPC code is generated to generate a codeword having a length of 8 because there are eight columns. Each column corresponds to eight encoded bits.

2 is a graph showing a Tanner graph corresponding to H1 in FIG.

2, the Tanner graph of the LDPC code includes eight variable nodes x ₁ (202), x ₂ (204), x ₃ (206), x ₄ (208), x ₅ (210) ₆ 212, x ₇ 214 and x ₈ 216 and four check nodes 218, 220, 222 and 224. Here, i th column and j th row in the parity check matrix H1 of the LDPC code correspond to each variable node x _i and the j th check node. The value of 1, that is, a value other than 0 at the intersection of the i-th column and the j-th row of the parity check matrix H1 of the LDPC code means that the variable nodes x _i and This means that there is an edge between the jth check node.

In the Tanner graph of the LDPC code, the degree of the variable node and the check node means the number of line segments connected to the respective nodes. This means that in the parity check matrix of the LDPC code, Is equal to the number of non-zero entries. For example, the variable nodes x ₁ 202, x ₂ 204, x ₃ 206, x ₄ 208, x ₅ 210, x ₆ 212, x ₇ 214), x ₈ (216) orders, respectively order 4, 3, 3, 3, 2, 2, 2, and 2, check nodes (218, 220, 222, 224) the order of the is, as each sequence of 6, 5, 5, 5. The number of non-zero elements in each column of the parity check matrix H1 of FIG. 1 corresponding to the variable nodes of FIG. 2 may be calculated using the above-described orders 4, 3, 3, 3, 2, 2, 2, and the number of non-zero elements in each row of the parity check matrix H1 of FIG. 1 corresponding to the check nodes of FIG. 2 is in the order of 6, 5, 5, It matches.

In order to express a degree distribution for a node of an LDPC code, the ratio of the number of variable nodes with degree _i to the total number of variable nodes is f _i , and the ratio of the number of check nodes with degree j to the total number of check nodes Let g _j . For example, for the LDPC code to the Figure correspond to the 1 and 2 is _{f 2 = 4/8, f} 3 = 3/8, f 4 = 1/8, i ≠ 2, 3, for a 4 f _i = 0, g ₅ = 3/4, g ₆ = 1/4, and g _j = 0 for j ≠ 5,6. Assuming that the length of the LDPC code is N, that is, the number of columns is N, and the number of rows is N / 2, the density of non-zero elements in the parity check matrix having the above- .

If N increases in Equation (1), the density of 1 in the parity check matrix continues to decrease. Generally, the LDPC code is inversely proportional to the nonzero element density with respect to the code length N, so that when N is large, the LDPC code has a very low density. The term low-density in the name of an LDPC code comes from this reason.

The characteristics of the parity check matrix of the LDPC code will now be described with reference to FIG. 3 shows an LDPC code adopted as a standard technique in DVB-S2, which is one of the European digital broadcasting standards.

은 LDPC 부호어의 길이이고,

은 정보어의 길이이고,

은 패리티의 길이를 의미한다. 그리고,

이 성립하도록 정수

과

를 결정한다. 이때,

도 정수가 되도록 한다. 편의상 도 3의 패리티 검사 행렬을 제 1패리티 검사 행렬

이라 한다. Referring to Figure 3,

Is the length of the LDPC codeword,

Is the length of the information word,

Is the length of the parity. And,

To establish this constant

and

. At this time,

Is also an integer. For convenience, the parity check matrix of FIG. 3 is referred to as a first parity check matrix

Quot;

번째 열(column)부터

번째 열까지의 구조는 이중 대각(dual diagonal) 형태이다. 따라서, 패리티 부분에 해당하는 열의 차수(degree) 분포는 그 값이 '1'인 마지막 열을 제외하고 모두 '2'를 가진다. Referring to FIG. 3, a portion corresponding to a parity portion in a parity check matrix,

From the first column

The structure up to the second column is a dual diagonal form. Therefore, the degree distribution of the column corresponding to the parity part has '2' except for the last column in which the value is '1'.

번째 열까지의 구조를 이루는 규칙은 다음과 같다. In the parity check matrix, the portion corresponding to the information word portion, i.e.,

The rules for the structure up to the first column are as follows.

개의 열을

개씩 그룹화(grouping)하여, 총

개의 열 그룹(column group)을 생성한다. 각 열 그룹에 속해있는 각각의 열을 구성하는 방법은 하기 규칙 2에 따른다. <Rule 1>: In the parity check matrix,

Columns

By grouping them one by one,

&Lt; / RTI &gt; column groups. The method for constructing each column belonging to each column group is as follows.

번째

열 그룹의 각 0번째 열의 1의 위치를 결정한다. 여기서, 각

번째 열 그룹의 0번째 열의 차수를

라 한다. 각 1이 있는 행의 위치를

이라 가정하면,

번째 열 그룹 내의

번째 열에서 1이 있는 행의 위치

는 하기 <수학식 2>와 같이 정의된다. <Rule 2>: First

th

The position of 1 in each 0th column of the column group is determined. Here,

The order of the 0th column in the ith column group is

. The position of the line with each 1

Assuming that,

Within the first column group

The position of the row with 1 in the third column

Is defined as < EMI ID = 2.0 >

,

,

.,

,

.,

,

번째

열 그룹 내에 속하는 열들의 차수는 모두

로 일정하다.According to the above rule

th

The order of the columns belonging to the column group is

.

 It is known that an LDPC code designed through the above rules can be efficiently encoded using a structural form. However, in the case of the DVB-S2 standard using the LDPC code, there are only two block lengths of the LDPC code used due to the limited use of the LDPC code, and in order to support the two block lengths It is necessary to store different parity check matrices.

Next, each step of the LDPC encoding process using the parity check matrix of the DVB-S2 as described above will be described.

인 정보어 비트들을

로 나타내고, 길이가

인 패리티 비트들을

로 나타낸다. 하기에서 구체적으로 설명하는 LDPC 부호는

,

,

,

의 특성을 가진다. First, for convenience of explanation,

Information bits

And the length is

Parity bits

Respectively. The LDPC code specifically described in the following

,

,

,

.

. Step 1 : Initialize the parity bits

.

Step 2: From the stored information of the parity check matrix, call the information of the row where 1 of the first column in the first column group of the information word is located:

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

.

,

,

.

를 이용하여 하기의 <수학식 3>과 같이 특정 패리티 비트

들을 업데이트한다. 여기서,

는 각각의

값 을 의미한다.The called information and information bits

As shown in Equation (3) below,

Lt; / RTI &gt; here,

Respectively,

Lt; / RTI >

, , ,

,,,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

.

.

는

로 표기하기도 할 수도 있고,

는 이진(binary) 덧셈을 의미한다.In Equation (3)

The

Or,

Means binary addition.

이후의 다음 359개의 정보어 비트

,

, ...,

에 대해서 먼저 하기의 <수학식 4>를 이용하여 구한다. Step 3:

The following 359 information bits

,

, ...,

Is first obtained using Equation (4) below.

, , .

,,.

는 각각의

값을 의미한다. 상기 <수학식 4>는 <수학식 2>와 동일한 개념의 수식임에 유의한다. In Equation (4)

Respectively,

Lt; / RTI &gt; Note that Equation (4) is an equation having the same concept as Equation (2).

에 대해서

를 업데이트한다. 예를 들어

, 즉,

에 대해서 하기의 <수학식 5>와 같이

들을 업데 이트 한다. Next, an operation similar to Equation (3) is performed using the value obtained from Equation (4). In other words,

about

Lt; / RTI &gt; E.g

, In other words,

As shown below in Equation (5)

.

임에 유의한다. 위와 같은 과정을

에 대해서 마찬가지로 진행한다. In the case of Equation (5)

. The above procedure

.

에 대해서

, (

)의 정보를 호출하고, 특정

를 업데이트한다. 여기서,

는

를 의미한다.

이후의 다음 359개의 정보어 비트

,

, ...,

에 대해서 <수학식 4>를 유사하게 적용하여

,

를 업데이트한다. Step 4: Like Step 2 above 361st information bit

about

, (

), And the specific

Lt; / RTI &gt; here,

The

.

The following 359 information bits

,

, ...,

Equation (4) is applied similarly to Equation

,

Lt; / RTI >

Step 5: Repeat steps 2, 3 and 4 for all 360 information bit groups. Finally, a parity bit is determined using Equation (6).

, .

,.

들이 LDPC 부호화가 완료된 패리티 비트들이다. In Equation (6)

Are parity bits for which LDPC coding is completed.

In the DVB-S2 as above, the encoding process proceeds from step 1 to step 5.

In order to apply the LDPC code as described above to an actual communication system, the LDPC code should be designed to be suitable for the data transmission amount required in the communication system. In particular, in a communication system supporting a variety of broadcasting services as well as an adaptive communication system employing a hybrid automatic retransmission request (HARQ) scheme and an adaptive modulation and coding (AMC) scheme, Accordingly, it is necessary to use an LDPC code having various block lengths in order to support various data transmission amounts.

However, in the case of the LDPC code used in the DVB-S2 system as described above, only two block lengths are required due to limited use, and independent parity check matrices are required. However, if the usable block length is limited as described above, it is difficult to increase the scalability and flexibility of the system.

In addition, storing all the parity check matrices corresponding to each block length of the LDPC code has a problem of deteriorating memory efficiency.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a channel coding / decoding apparatus and a control method thereof in a communication system using LDPC codes of various block lengths.

It is another object of the present invention to provide a method and apparatus for generating an LDPC code having various block lengths by using a given LDPC code without designing a new parity check matrix in order to increase memory efficiency.

According to another aspect of the present invention, there is provided a method of encoding a structured LDPC code having a variable length, the method comprising the steps of: detecting first parity check matrix information stored in advance; determining a block length of a required LDPC code Determining a size to group information words of a second parity check matrix; constructing a partial matrix corresponding to an information word of the second parity check matrix to correspond to the grouping size from the first parity check matrix; Constructing a partial matrix corresponding to a parity of the second parity check matrix and performing encoding based on the first parity check matrix or the second parity check matrix, .

In an apparatus for encoding an LDPC code having a variable length, after checking a block length of a required LDPC code, a size to group the information words of the second parity check matrix is determined, and a first parity A partial matrix corresponding to an information word of the second parity check matrix and a partial matrix corresponding to a parity corresponding to the grouping size from the check matrix and forming one of the first parity check matrix and the second parity check matrix A modulator for modulating the LDPC code according to a predetermined modulation scheme to generate a modulation symbol, and a transmitter for transmitting the modulation symbol.

There is provided a method of decoding a structured LDPC code having a variable length, the method comprising the steps of: receiving a signal; detecting first stored parity check matrix information; determining a block length of an LDPC code to be decoded; Determining a size to group information words of a second parity check matrix; and constructing a partial matrix corresponding to an information word of the second parity check matrix from the first parity check matrix to correspond to the grouping size And generating a partial matrix corresponding to a parity of the second parity check matrix by decoding the received signal based on one of the first parity check matrix and the second parity check matrix, The method includes the steps of:

The present invention relates to an apparatus for decoding an LDPC code having a variable length, the apparatus comprising: a receiver for receiving a signal; a modulator for demodulating and outputting the received signal by a predetermined demodulation scheme; The size of the information word of the second parity check matrix is determined, and the information of the second parity check matrix is calculated from the first parity check matrix stored beforehand, And a decoder for decoding the received signal based on one of the first parity check matrix and the second parity check matrix to detect the partial matrix corresponding to the partial matrix and the parity corresponding to the word by using the LDPC code .

In the present invention that operates as described in detail above, the effects obtained by the representative ones of the disclosed inventions will be briefly described as follows.
The present invention has an advantage of increasing the scalability and flexibility of a system by generating an LDPC code having various block lengths by using information of a given parity check matrix in a communication system using an LDPC code.

Also, the present invention is advantageous in that the efficiency of the memory is increased because only one parity check matrix information can be stored even in a system supporting various block lengths.

Preferred embodiments according to the present invention will be described in detail with reference to the accompanying drawings. In the following description, only parts necessary for understanding the operation according to the present invention will be described, and the description of other parts will be omitted so as not to obscure the gist of the present invention.

The present invention generates an LDPC code having various block lengths larger or smaller than the block length by using a parity check matrix of one LDPC code having a specific form and encodes and decodes the LDPC code using the LDPC codes having various block lengths And an apparatus and a method for providing the same.

4 is a block diagram of a communication system using an LDPC code.

는 전송되기 전에 송신기(410)의 LDPC 부호화기(encoder)(411)를 통해 부호화되고, 변조기(Modulator)(413)에 의해 변조되어 무선 채널(420)을 통해 전송된다. 그러면, 수신기(430)의 복조기(Demodulator)(430)는 상기 무선 채널을 통해 전송되는 신호 r을 수신하여 복조하고, LDPC 복호기(Decoder)(433)는 상기 복조된 신호로부터 메시지의 추정치(estimate)

를 추정해낸다.Referring to Figure 4,

Encoded by the LDPC encoder 411 of the transmitter 410 before being transmitted and modulated by the modulator 413 and transmitted via the wireless channel 420. [ The demodulator 430 of the receiver 430 receives and demodulates a signal r transmitted through the wireless channel and an LDPC decoder 433 estimates a message from the demodulated signal.

.

The LDPC encoder 411 and the LDPC decoder 433 generate a parity check matrix corresponding to a block length required in the communication system from a predetermined scheme. Particularly, in the present invention, the LDPC encoder 411 and the LDPC decoder 433 can support various block lengths without the need for additional storage information using LDPC codes. The generation of the parity check matrix corresponding to the block length required in the communication system may be performed in the LDPC encoder 411 and the LDPC decoder 433 as described above, The detailed operation of the LDPC encoder 411 and the LDPC decoder 433 for supporting various block lengths will be described in detail with reference to the following description of a coding method.

A structured LDPC coding method having a variable length according to the present invention will now be described.

The encoding method of the present invention detects a stored first parity check matrix information, generates a second parity check matrix corresponding to a block length of a required LDPC code, and generates the first parity check matrix or the newly generated second parity check matrix And performing encoding based on one of the check matrices.

 First, a process of generating a parity check matrix corresponding to a variable block length from a stored parity check matrix will be described.

개의 열 단위로 그룹화하는 과정과, 각 열 그룹의 첫 번째 열의 차수와 1의 위치에 의해 그모든 특성이 결정된다. 따라서, 상술한 바와 같은 특성을 갖는 LDPC 부호로부터 다른 블록 길이를 가지는 LDPC 부호를 지원하기 위해서는 주어진 패리티 검사 행렬로부터 새롭게 그룹화하는 과정과, 상기 그룹화 과정을 통해서 새롭게 얻어진 열 그룹의 첫 번째 열의 1의 위치만 결정하면 충분하다. According to the conventional design rule of the parity check matrix, the structure of the LDPC code is

All the characteristics are determined by the process of grouping in units of columns and the order of the first column and the position of 1 in each column group. Therefore, in order to support LDPC codes having different block lengths from the LDPC codes having the above-described characteristics, a new grouping process is performed from a given parity check matrix. It is enough to decide.

The present invention proposes a method of supporting an LDPC code having a different block length from an LDPC code having a specific block length.

, 정보어의 길이는

, 패리티의 길이는

라 하자. 여기서

와

는 각각

과

의 약수이며,

를 만족한다. 새롭게 설계한 제 2 LDPC 부호의 패리티 검사 행렬을 제 2 패리티 검사 행렬

라 하자. 5 is a flowchart for generating an LDPC code having a different block length from a parity check matrix of an LDPC code stored according to an embodiment of the present invention.
For convenience of explanation, the length of the codeword of the second LDPC code to be newly supported from the stored first LDPC code is

, The length of the information word is

, The length of the parity is

Let's say. here

Wow

Respectively

and

And,

. The parity check matrix of the newly designed second LDPC code is referred to as a second parity check matrix

Let's say.

을 검출해낸다. 그리고, 520 단계에서 LDPC 부호화기는 하기의 <규칙 3>에서와 같이 필요한 제 2 LDPC 부호의 블록 길이를 확인한 후, 그에 상응하는

를 결정한다.
Referring to FIG. 5, in step 510, the LDPC encoder calculates a first parity check matrix

. Then, in step 520, the LDPC encoder checks the block length of the second LDPC code necessary in < Rule 3 >

.

delete

로 정의한다. 패리티 검사 행렬에서 정보어에 해당하는

개의 열을

개씩 그룹화하여 총

개의 열 그룹을 만든다.

의 값은 상기 규칙 1과 2에서 설명한 제 1 패리티 검사 행렬

에 대한 열 그룹의 개수

와 동일한 수가 되며, 하기의 <수학식 7>과 같이

또한 동일함을 알 수 있다. . <Rule 3>:

. In the parity check matrix,

Columns

Group by group

Create a group of columns.

The values of the first parity check matrix < RTI ID = 0.0 >

Number of column groups for

And as shown in Equation (7) below,

The same is also true. .

.

.

Then, in step 530, the LDPC encoder constructs a partial matrix corresponding to the information word in the parity check matrix of the LDPC code by applying the following < Rule 4 >.

에서의 1의 위치와 상기

및

를 이용하여 새롭게 생성할 패리티 검사 행렬

에서의

번째 열 그룹 내의

번째 열에서 1이 있는 행의 위치

는 하기의 <수학식 8>과 같이 정의된다. &Lt; Rule 4 >< EMI ID =

1 &lt; / RTI &gt;

And

A parity check matrix to be newly generated

In

Within the first column group

The position of the row with 1 in the third column

Is defined as Equation (8) below.

와 에 대해서,

And,

,

,

,

,

.

.

In step 540, the LDPC encoder constructs a partial matrix corresponding to the parity in the parity check matrix of the LDPC code by applying the following < Rule 5 >.

번째 열부터

번째 열까지의 구조는 이중 대각(dual diagonal) 형태가 되도록 설정한다. 즉, 패리티 부분에 해당하는 열의 차수 분포는 마지막 열이 1임을 제외하고 모두 2를 가진다. &Lt; Rule 5 &gt;: A part corresponding to the parity part in the parity check matrix,

From the first column

The structure up to the second column is set to be a dual diagonal form. That is, the degree distribution of the column corresponding to the parity part has 2 except the last column is 1.

는 종래의 <규칙 1>, <규칙 2>를 이용하여 만든 제1 패리티 검사 행렬

의 정보를 그대로 사용하여 설계하였기 때문에

의 저장을 위해 별도의 메모리가 필요하지 않다. 즉,

의 정보만 가지고 있으면

를 생성한 후 LDPC 부호화기에 적용하여 LDPC 부호어를 생성할 수 있다. As described above, the second parity check matrix generated using <Rule 3>, <Rule 4> and <Rule 5>

A first parity check matrix < RTI ID = 0.0 >

Because it was designed using the information of

No additional memory is required for storage. In other words,

If you only have the information

And then applies it to the LDPC encoder to generate the LDPC codeword.

Next, the encoding process of a new second LDPC code performed through the method proposed by the present invention will be described in detail.

, 정보어의 길이가

인 예를 들어 설명하기로 한다. 이에 따라

이므로 상기

의 정의에 의해서

이 되며,

으로 고정된다. Here, for convenience of description, the block length for a new LDPC code to be generated is

, The length of the information word is

For example. Accordingly

Therefore,

By definition of

Lt; / RTI &

.

)한다. Step 1 : Initialize the parity bits (

)do.

)의 정보로부터 하기와 같은 첫 번째 열 그룹 내에서 첫 번째 열의 1이 위치한 행의 정보를 호출한다. Step 2 : The stored parity check matrix (

) Of the first column in the first column group as shown below.

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

.

,

,

.

,

로부터 다음과 같이

값을 구한다. Step 3: Applying rule 4 above

,

From

Value.

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

.

,

,

,

,

,

,

,

.

를 이용하여 다음 <수학식 9>와 같이 특정

들을 업데이트 한다. (여기서

는 각각의

값을 의미한다.)The information

As shown in the following Equation (9)

Lt; / RTI &gt; (here

Respectively,

Value.)

, , ,

,

,

,

,,,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

. (

는 이진(binary) 덧셈을 의미함.)

. (

Means binary addition.)

delete

이후의 다음 89개의 정보어 비트

,

, ...,

에 대해서 먼저 다음 <수학식 10>을 구한다. Step 4:

The next 89 information bits

,

, ...,

The following Equation (10) is obtained.

,, .

,,.

는 각각의

값을 의미한다. 상기 <수학식 10>은 <수학식 2>와 동일한 개념의 수식임에 유의한다. In Equation (10)

Respectively,

Lt; / RTI &gt; Note that Equation (10) is a formula having the same concept as Equation (2).

에 대해서

를 업데이트한다. 예를 들어

, 즉,

에 대해서 다음 <수학식 11>과 같이 특정

들을 업데이트 한다. Next, an operation such as Equation (9) is performed using the value obtained from Equation (10). In other words,

about

Lt; / RTI &gt; E.g

, In other words,

As shown in the following Equation (11)

Lt; / RTI >

, , ,

,,,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

. (

는 이진(binary) 덧셈을 의미함.)

. (

Means binary addition.)

임에 유의한다. 위와 같은 과정을

에 대해서 마찬가지로 진행한다. In the case of Equation (11)

. The above procedure

.

에 대해서

, (

)의 정보를 호출하고

를 계산하고 특정

를 업데이트 한다. (여기서

는

를 의미한다.)

이후의 다음 89개의 정보어 비트

,

, ...,

에 대해서 <수학식 10>을 적용하여

,

들을 업데이트한다. 90개의 정보어 비트 그룹의 각각에 대해서 상기 단계 2, 3, 4의 과정을 반복한다. Step 5: As in step 3 above,

about

, (

) &Lt; / RTI >

And calculates

Lt; / RTI &gt; (here

The

.)

The next 89 information bits

,

, ...,

(10) &lt; RTI ID = 0.0 &gt;

,

Lt; / RTI &gt; Repeat steps 2, 3, and 4 for each of the 90 information word groups.

Step 6: Finally, a parity bit is determined through Equation (12).

, .

,.

들이 LDPC 부호화가 완료된 패리티 비트들이다. In Equation (12)

Are parity bits for which LDPC coding is completed.

로부터

를 구하는 과정을 살펴보면

이므로, 하기의 <수학식 13>이 성립하게 된다. On the other hand, in step 3

from

The process of obtaining

The following Equation (13) is established.

를 구하는 별도의 과정 없이 다음과 같이 하나의 단계로 정리하여 표현할 수도 있다. Thus, steps 3 , 4, and 5

Can be summarized in one step as shown below without any additional procedure for obtaining.

Combination of steps 3, 4 and 5 :

,

,

에 대해서 <수학식 14>와 같이

를 업데이트한다.

,

,

As shown in Equation (14)

Lt; / RTI &gt;

,

이다. In Equation (14)

,

to be.

The embodiment of the LDPC coding scheme described above is a very limited case, and various parity check matrices can be generated based on the generation rule of the parity check matrix of the LDPC code.

The coding scheme is a method in which shortening and puncturing are not applied to a second LDPC code to be newly generated from a first parity check matrix. In the second parity check matrix, a code rate, The algebraic characteristic of the code such as the order distribution of the first parity check matrix is the same as the first parity check matrix. Therefore, in order to obtain codes different in sign algebraic characteristic from the first parity check matrix, there is a method of applying the shortening to the generated second parity check matrix and then applying puncturing to a part of the LDPC codeword obtained by applying the encoding method.

A method of obtaining LDPC codes having different logarithmic characteristics by applying shortening and puncturing to a second parity check matrix obtained by using the above-described <Rule 3>, <Rule 4>, <Rule 5>, etc. from the first parity check matrix I will explain in detail.

,

,

이라 하자. 그러면 <규칙 3>, <규칙 4>, <규칙 5>에 의거하여 제 1 패리티 검사행렬로부터 블록 길이, 정보어 길이 및 그룹화 크기가

,

,

인 제 2 패리티 검사행렬을 얻을 수 있었다. 이 때 새롭게 생성된 상기 제 2 패리티 검사행렬 자체는 부호율이나 열과 행의 차수 분포 등과 같은 부호의 대수적 특성이 제 1 패리티 검사행렬과 동일함에 유의한다.For convenience of explanation, the block length, information length, and grouping size of the first parity check matrix corresponding to the parity check matrix of the LDPC code shown in <Rule 1> and <Rule 2>

,

,

. Then, the block length, the information word length, and the grouping size are calculated from the first parity check matrix based on <Rule 3>, <Rule 4>, and <Rule 5>

,

,

A second parity check matrix is obtained. It should be noted that the newly generated second parity check matrix itself has the same algebraic characteristics as the first parity check matrix, such as a coding rate, a degree distribution of a column and a row.

,

하자. 만일

,

라고 정의하면, 제 2 패리티 검사행렬에서

비트만큼 단축을 취하고,

비트만큼 천공을 취하면 블록 길이와 정보어 길이를 각각

,

인 제3 LDPC 부호를 얻을 수 있다. 이렇게 생성된 제3 LDPC 부호는

또는

일 때, 부호율이

가 되어 일반적으로 제 2 패리티 검사행렬의 부호율

와는 다르게 되므로 대수적 특성이 변하게 된다.

인 경우에는 단축이나 천공을 적용하지 않는 경우에 해당된다. Now, the block length and the information length of the third LDPC code to be finally obtained are respectively

,

lets do it. if

,

, The second parity check matrix

Taking the bit as short as possible,

If the puncturing is performed by a bit, the block length and the information word length are respectively

,

The third LDPC code can be obtained. The thus generated third LDPC code

or

, The code rate is

The coding rate of the second parity check matrix

And the algebraic characteristic is changed.

In the case of the case where no shortening or perforation is applied.

,

,

,

의 특성을 가지는 LDPC 부호를 이용하여 단축과 천공을 통하여 블록 길이가

이며, 정보어의 길이가

인 새로운 제3 LDPC 부호를 생성하는 과정의 그 구체적인 예를 살펴보기로 한다.Then, referring to FIG. 6, as a first parity check matrix

,

,

,

LDPC code is used to shorten and puncture the block length

And the length of the information word is

Hereinafter, a specific example of a process of generating a new third LDPC code will be described.

을 검출해낸다. 그리고, 620 단계에서 LDPC 부호화기는 상기의 <규칙 3>에서와 같이 필요한 LDPC 부호의 블록 길이를 확인한 후, 그에 상응하는 그룹화 크기

를 결정한다. 상기 620 단계에 의해 <규칙 3>에 의해

로 설정하고, 630 단계에서 단축과 천공의 필요 여부를 확인한다. 그리고, 우선 640 단계 및 650 단계에서 <규칙 4>, <규칙 5>로부터

,

인 제2LDPC 부호의 제 2 패리티 검사행렬을 생성할 수 있다. 이때 ,

,

이 성립하므로, 660 단계에서 제 2 패리티 검사행렬에 대해 1320 비트만큼 단축을 취하고 30 비트만큼 천공을 취하면 블록 길이가

이며, 정보어의 길이가

인 새로운 제3LDPC 부호를 생성할 수 있다. 참고로 상기 제 1 패리티 검사 행렬의 부호율은 4/9이며, 상기 제 2 패리티 검사행렬의 부호율은 4/15로서 서로 상이함을 알 수 있다. Referring to FIG. 6, in step 610, the LDPC encoder calculates a first parity check matrix

. Then, in step 620, the LDPC encoder checks the block length of the required LDPC code as in the above < Rule 3 &gt;, and then,

. In step 620, by < Rule 3 >

In step 630, it is determined whether or not there is a need for shortening and drilling. First, in steps 640 and 650, < Rule 4 > and < Rule 5 >

,

The second parity check matrix of the second LDPC code can be generated. At this time ,

,

Therefore, if the second parity check matrix is shortened by 1320 bits and punctured by 30 bits in step 660,

And the length of the information word is

Lt; RTI ID = 0.0 &gt; LDPC &lt; / RTI &gt; The coding rate of the first parity check matrix is 4/9 and the coding rate of the second parity check matrix is 4/15, which are different from each other.

As a shortening method for the second parity check matrix, there may be various methods. One concrete method is as follows.

)열 그룹의 각 0번째 열의 1의 위치에 의해서 해당 열 그룹의 모든 열의 1의 위치를 얻을 수 있다.

,

,

,

의 특성을 가지는 LDPC 부호는 총 20개의 정보어 열 그룹에 대한 정보가 주어져 있으며, i 번째 (i = 1,2,....,20)열 그룹의 각 0번째 열의 1의 위치 정보는 다음과 같다: According to <Rule 2>, the i-th (i = 1, ...,

) The position of 1 in every column of the column group can be obtained by the position of 1 in each 0th column of the column group.

,

,

,

(I = 1, 2, ..., 20), the position information of the 1 &lt; th &gt; Is as follows:

일 때, 정보어의 길이는

이 된다. 하지만, 위의 열 그룹에 대한 정보 중에서 다음과 같이 9 개의 정보어 열 그룹에 대한 정보만 사용한다고 가정하자. If all of the information on the information word group is used as it is, it is determined by the rule 3, the rule 4, and the rule 5

, The length of the information word is

. However, suppose that the information on the above column group is used only for the nine information word groups as follows.

에 대해 패리티 검사행렬을 생성하면, 정보어의 길이는

이 되어, 사용하지 않은 11 개의 정보어 열 그룹에 대한 단축을 적용한 것과 동일한 효과를 얻을 수 있다. By using only the information on the nine information word groups, it is possible to search for the information word groups by using <Rule 3>, <Rule 4>, and <Rule 5>

, The length of the information word is

So that it is possible to obtain the same effect as applying the shortening to the unused 11 information word group.

The example of the shortening is only an example of shortening for a column group, and various shortening methods such as shortening by bit can be applied.

As a typical example of the puncturing method, there is a method of puncturing 1 bit per parity bit when the number of puncturing bits is denoted by the length of the parity of the LDPC code corresponding to the newly generated second parity check matrix. Various methods can be applied.

1 is a diagram illustrating an example of a parity check matrix of an LDPC code having a general length of 8,

2 shows a Tanner graph of an example of a parity check matrix of an LDPC code having a general length of 8,

FIG. 3 is a schematic diagram of a general DVB-S2 LDPC code,

4 is a block diagram of a digital communication system using an LDPC code according to the present invention.

5 is a flowchart for generating an LDPC code having a different block length from a parity check matrix of a stored LDPC code according to an embodiment of the present invention.

6 is a flowchart for generating an LDPC code having a different block length by combining shortening and puncturing from a parity check matrix of a stored LDPC code according to another embodiment of the present invention.

Claims (14)

A method for coding a structured LDPC code having a variable length in an encoder, Detecting first parity check matrix information stored in the first parity check matrix; Determining a size of a group of information words of a second parity check matrix after checking a block length of a required LDPC code; Constructing a partial matrix corresponding to an information word of the second parity check matrix from the first parity check matrix to correspond to the grouping size; Constructing a partial matrix corresponding to a parity of the second parity check matrix; And performing encoding based on one of the first parity check matrix and the second parity check matrix. The method according to claim 1, Wherein a size of grouping information words of the second parity check matrix is a divisor or a multiple of a grouping size of information words of the first parity check matrix. The method according to claim 1, The process of constructing a partial matrix corresponding to the information word includes: Determining a position of 1 of the information word in the second parity check matrix using a position of 1 of the information word in the first parity check matrix and a size of grouping information words of the second parity check matrix; Wherein the coding method comprises the steps of: 3. The method of claim 2, The process of constructing a partial matrix corresponding to the information word includes: In the second parity check matrix,

Within the first column group

The position of the row with 1 in the third column

Is defined as Equation (13) below. &Lt; EMI ID = 13.0 > &Quot; (13) &quot;

Wow

about,

,

,

. The method according to claim 1, Wherein the step of constructing the partial matrix corresponding to the parity of the second parity check matrix comprises: And constructing the partial matrix to correspond to the grouping size. The method according to claim 1, And applying shortening and puncturing to the second parity check matrix. An apparatus for coding an LDPC code having a variable length, Determining a size of a grouping of the information words of the second parity check matrix after checking the block length of the required LDPC code and determining a size of the group of information words of the second parity check matrix based on the stored first parity check matrix, A partial matrix corresponding to the information word and a partial matrix corresponding to the parity are constructed, An encoder for performing encoding based on one of the first parity check matrix and the second parity check matrix; A modulator for modulating the LDPC code with a predetermined modulation scheme to generate a modulation symbol, And a transmitter for transmitting the modulation symbols. 8. The method of claim 7, Wherein a size of grouping the information words of the second parity check matrix is a divisor or a multiple of a grouping information word of the first parity check matrix. 8. The method of claim 7, The encoder determines the position of 1 of the information word in the second parity check matrix using the position of 1 of the information word in the first parity check matrix and the size of the information word of the second parity check matrix, Wherein the encoding unit comprises: 10. The method of claim 9, Wherein the first parity check matrix

Within the first column group

The position of the row with 1 in the third column

Is defined as Equation (14) below. &Lt; EMI ID = 14.0 > &Quot; (13) &quot;

Wow

about,

,

,

.  8. The method of claim 7, Wherein the encoder is configured to correspond to the grouping size of the partial matrix. 8. The method of claim 7, Wherein the encoder applies shortening and puncturing to the second parity check matrix. A method for decoding a structured LDPC code having a variable length in a decoding apparatus, Receiving a signal, Detecting first parity check matrix information stored in the first parity check matrix; Determining a size of a grouping of information words of a second parity check matrix after confirming a block length of an LDPC code to be decoded; Constructing a partial matrix corresponding to an information word of the second parity check matrix from the first parity check matrix to correspond to the grouping size; Constructing a partial matrix corresponding to a parity of the second parity check matrix; And decoding the received signal based on one of the first parity check matrix and the second parity check matrix to detect the received signal as the LDPC code. An apparatus for decoding an LDPC code having a variable length, A receiver for receiving a signal, A demodulator for demodulating the received signal by a predetermined demodulation scheme and outputting the demodulated signal, After determining the length of the LDPC code to be decoded for the signal output from the modulator, the size of the group of information words of the second parity check matrix is determined, A partial matrix corresponding to an information word of the second parity check matrix and a partial matrix corresponding to a parity corresponding to the grouping size from a first parity check matrix stored in advance, And a decoder for decoding the received signal based on one of the two parity check matrices and detecting the received signal as the LDPC code.

Images