KR20110055410A - Method and apparatus for transmitting and receiving data in communication system - Google Patents

Abstract

PURPOSE: A data transmission/reception apparatus and method thereof in a communication system are provided to increase diversity gain by transmitting parity bits through a plurality of frames and to minimize additional module in a coding/decoding process. CONSTITUTION: A control unit(1604) controls the determination of reduction target bit number according to the length of information words in a reduction application unit. The reduction application unit removes a raw corresponding to the reduced bit in a parity inspection matrix of an LDPC code. An LDPC encoder(1602) performs coding. An additional parity bit group generator(1608) generates additional parity bit groups. A frame configuration unit(1610) configures a frame.

Description

METHOD AND APPARATUS FOR TRANSMITTING AND RECEIVING DATA IN COMMUNICATION SYSTEM}

The present invention relates to a communication system, and more particularly, to a transmission and reception method and apparatus for achieving diversity effect.

In a wireless communication system, the performance of a link is significantly degraded due to various noises, fading, and inter-symbol interference (ISI) of a channel. Therefore, in order to implement high-speed digital communication systems requiring high data throughput and reliability such as next-generation mobile communication, digital broadcasting, and portable Internet, it is essential to develop a technique for overcoming noise, fading, and ISI. As part of such research, recently, researches on error-correcting codes have been actively conducted as a method for increasing the reliability of communication by efficiently restoring information distortion.

Originally introduced by Gallager in the 1960s, the LDPC code has long been forgotten due to the complexity of implementation far beyond the technology of the time. However, as the turbo code found by Berrou, Glavieux, and Thitimajshima in 1993 showed performance close to Shannon's channel capacity, it was repeatedly decoded with much interpretation of the performance and characteristics of the turbo code. Much research has been done on iterative decoding and channel coding based on graphs. In the late 1990s, the LDPC code was re-studied and iterative decoding based on a sum-product algorithm on a Tanner graph (a special case of a factor graph) corresponding to the LDPC code was applied. It has been found that the performance is close to Shannon's channel capacity.

개의 비트 혹은 심볼들로 구성되어 있는 정보어를 입력 받아 LDPC 부호화를 하여

개의 비트 혹은 심볼들로 구성되어 있는 부호어를 생성한다. 이하에서는 설명의 편의를 위하여

개의 비트들로 구성되어 있는 정보어를 입력 받아

개의 비트들로 구성되어 있는 부호어에 대하여서만 설명하고자 한다. 즉,

개의 입력 비트들인 정보어

를 입력 받아 LDPC 부호화를 통해

를 생성한다. 이때 systematic 부호인 경우

로 구성되며,

는 패리티 비트들이라 하고 패리티 비트들의 개수 이다.The LDPC code is generally defined as a parity-check matrix and can be expressed using a bipartite graph collectively referred to as a Tanner graph. LDPC code is

LDPC encoding by receiving an information word consisting of four bits or symbols

Generates a codeword consisting of four bits or symbols. Hereinafter, for convenience of explanation

Receives an information word consisting of four bits

Only a codeword consisting of two bits will be described. In other words,

Information bits

Through the LDPC encoding

. In the case of systematic sign

Lt; / RTI >

Are parity bits and the number of parity bits to be.

가 부호어가 됨을 의미한다. Meaning that it is defined by the parity check matrix, LDPC code is a sequence that satisfies Equation 1

Means to be a codeword.

는

개의 열(column)들로 구성되어 있으며 i 번째 열은 i 번째 부호어 비트

와 연관 관계가 있음을 의미한다. The parity check matrix

Is

I columns consists of i columns

It means that there is an association with.

개의 변수 노드(variable node)와 검사 노드(check node)라 불리는 정점들로 이루어진 이분 그래프로 표현된다. 상기 변수 노드는 부호화된 비트와 일대일 대응된다. 즉 i 번째 변수 노드는 i 번째 부호어 비트

에 대응됨을 의미한다. In addition, the LDPC code may be represented by a bipartite graph based on a parity check matrix. The bipartite graph means that the vertices constituting the graph are divided into two different types. In the case of the LDPC code,

It is represented by a bipartite graph of two variable nodes and vertices called check nodes. The variable node corresponds one-to-one with the coded bits. That is, the i th variable node is the i th codeword bit.

Corresponding to.

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

1 is an example of the parity check matrix H of the LDPC code consisting of four rows and eight columns. Referring to FIG. 1, since the parity check matrix H has eight columns, the parity check matrix H generates an LDPC codeword having a length of eight, and each column corresponds to the encoded eight bits.

FIG. 2 is a diagram illustrating a Tanner graph corresponding to H of FIG. 1.

Referring to 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), x ₆ (212), x ₇ (214), x ₈ (216) and four check nodes (218, 220, 222, 224). Here, the i th column and the j th row of the parity check matrix H of the LDPC code correspond to the i th variable node x _i and the j th check node, respectively. In addition, a value of 1, i.e., a non-zero value of a point where the i-th column and the j-th row of the parity check matrix H of the LDPC code intersect, may mean the non-zero value of the variable node x _i on the Tanner graph as shown in FIG. An edge exists between the j th test nodes.

In general, LDPC codes are decoded using a Tanner graph. That is, a variable node and a check node generate a message and perform repeated decoding while exchanging the message through each segment. Therefore, an association is formed between the variable nodes connected to one test node, and the association must be taken into consideration when shortening and drilling.

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 each node, which is determined in the column or row corresponding to the node in the parity check matrix of the LDPC code. It is equal to the number of nonzero entries. For example, in FIG. 2, the variable nodes x ₁ 202, x ₂ 204, x ₃ 206, x ₄ 208, x ₅ 210, x ₆ 212, and x ₇ ( 214, and the order of x ₈ (216) is 4, 3, 3, 3, 2, 2, 2, 2, respectively, in order, and the order of test nodes 218, 220, 222, and 224, respectively, in order. 6, 5, 5, 5. In addition, the number of non-zero elements in each column of the parity check matrix H of FIG. 1 corresponding to the variable nodes of FIG. 2 is equal to the above-described orders 4, 3, 3, 3, 2, 2, 2, The number of nonzero elements in each row of the parity check matrix H of FIG. 1 corresponding to the check nodes of FIG. 2 corresponding to the check nodes of FIG. 2 is in the order of the above orders 6, 5, 5, and 5. Matches as The variable nodes have a one-to-one correspondence with codeword bits. Therefore, if the i-th variable node and the i-th codeword bit have a one-to-one correspondence, the order of the i-th variable node can be regarded as the order of the i-th codeword bit.

이 증가하게 되면 패리티 검사 행렬 내에서 1의 밀도는 계속해서 감소하게 된다. 일반적으로 LDPC 부호는 부호 길이

에 대하여 0이 아닌 원소의 밀도가 반비례하므로,

이 큰 경우에는 매우 낮은 밀도를 가지게 된다. LDPC 부호의 명칭에서 저밀도(low-density)란 말은 이와 같은 이유로 유래되었다.

As this increases, the density of 1 in the parity check matrix continues to decrease. Typically LDPC codes are code length

Since the density of nonzero elements is inversely proportional to,

This large case has a very low density. The term low-density in the name of the LDPC code is derived for this reason.

Since the LDPC code is defined by the parity check matrix, the parity check matrix must be stored in order to apply the LDPC code in a given system. In general, to store the LDPC code, position information of weight 1 is stored in a parity check matrix. However, since the codeword length of the LDPC code used in the actual system ranges from hundreds to hundreds of thousands, if the codeword length of the LDPC code is very long, the memory required to store the location information of weight 1 is very large. There is this.

In order to overcome this drawback, various LDPC codes with specific structures have been studied. The LDPC code having a specific structure can store the position of the weight 1 very efficiently because the position of the weight 1 in the parity check matrix is limited according to a specific condition.

3 shows an example of an LDPC code having the specific structure. The LDPC code is a systematic structure in which a codeword includes an information word. Hereinafter, for convenience of description, the description will be made based on the parity check matrix of FIG. 3, but the LDPC code of the present invention is not limited to the parity check matrix of FIG. 3.

개의 열(column)로 구성되며, 패리티 파트는

개의 열로 구성된다. 상기 패리티 검사 행렬의 행(row)의 개수는 패리티 파트의 열의 개수와 동일한

로 구성된다. Referring to FIG. 3, the parity check matrix includes an information word part and a parity part, and the information word part is

Consists of two columns, the parity part

It consists of two columns. The number of rows of the parity check matrix is equal to the number of columns of the parity part.

It consists of.

은 LDPC 부호어의 길이,

은 정보어의 길이,

은 패리티 파트의 길이를 의미한다. 상기 부호어의 길이라 함은 부호어를 구성하는 비트들의 개수를 의미한다. 또한 정보어의 길이라 함은 정보어를 구성하는 비트들의 길이를 의미한다. 그리고

이 성립하도록 정수

과

를 결정한다. 이때,

도 정수가 되도록 한다. 상기

과

는 부호어 길이와 부호율에 따라 달라질 수 있다.

Is the length of the LDPC codeword,

Is the length of the information word,

Means the length of the parity part. The length of the codeword means the number of bits constituting the codeword. In addition, the length of the information word means the length of the bits constituting the information word. And

Integer to make

and

. At this time,

Is also an integer. remind

and

May vary according to codeword length and code rate.

번째 열(column)부터

번째 열까지의 무게 1의 위치는 이중 대각(dual diagonal) 구조를 가진다. 따라서 상기 패리티 비트에 대응되는 열의 차수(degree)는 상기

번째 열을 제외하고 모두 2이며, 상기

번째 열의 차수는 1 임을 알 수 있다. The part corresponding to the parity bit in the parity check matrix of FIG.

From the first column

The position of weight 1 up to the first row has a dual diagonal structure. Therefore, the degree of the column corresponding to the parity bit is

2 except for the 1st column, and above

It can be seen that the order of the first column is 1.

번째 열까지의 구조를 이루는 규칙은 다음과 같다. Referring to FIG. 3, the part corresponding to the information word in the parity check matrix, that is, from the 0th column

The rules forming 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 속한 각각의 열을 구성하는 방법은 하기 규칙 2를 따른다. <Rule 1>: Corresponds to the information word in the parity check matrix

Columns

Grouped one by one,

Create two column groups. How to configure each column belonging to each column group follows the rule 2.

번째

열 그룹의 각 0 번째 열에서의 1의 위치를 결정한다. 여기서, 각

번째 열 그룹의 0 번째 열의 차수를

라 할 때, 1이 있는 각 행의 위치를

이라 가정하면,

번째 열 그룹 내의

번째 열에서 1이 있는 행의 위치

는 하기 수학식 2와 같이 정의된다. <Rule 2>: First

th

Determine the position of 1 in each 0th column of the row group. Where

Order of the 0th column of the 1st column group

Is the position of each row

Assuming that

Within the first column group

Position of row with 1 in the first column

Is defined as in Equation 2 below.

번째

열 그룹 내에 속하는 열들의 차수는 모두

로 일정함을 알 수 있다. 이하에서는 상기 규칙에 따라 패리티 검사 행렬에 대한 정보를 저장하고 있는 LDPC 부호의 구조를 쉽게 이해하기 위하여 구체적인 예를 살펴본다. According to rule 1 and rule 2 above

th

The orders of the columns within a column group are all

It can be seen that the constant. Hereinafter, a specific example will be described to easily understand the structure of the LDPC code that stores information about the parity check matrix according to the above rule.

이며, 3개의 열 그룹의 0 번째 열에 대한 1이 있는 행의 위치 정보는 다음과 같은 수열들로 나타낼 수 있다. 여기서 이 수열들을 "무게-1 위치 수열(weight-1 position sequence)"이라 부른다. As a specific example

The positional information of a row having 1 for the 0th column of the 3 column group may be represented by the following sequences. These sequences are referred to herein as "weight-1 position sequences."

The weight-1 position sequence for the position of the row 1 with respect to the 0th column of each column group may be expressed only as a sequence for each column group as follows for convenience.

1 2 8 10

0 9 13

0 14

번째 무게-1 위치 수열은

번째 열 그룹에 대한 1이 있는 행의 위치 정보를 순차적으로 나타낸 것이다. That is

First weight-1 position sequence

Positional information of a row with 1 for the first column group is sequentially displayed.

와 부호어 비트수

에 대하여 부호화 및 복호화를 한다. 이때 패리티 비트수

는

이다. 만약 부호화기에 입력되는 정보어 비트수

보다 상기 주어진 정보어 비트수

가 클 경우에는

만큼 단축을 시켜서 전송한다. 또한 필요한 패리티 비트수

가 상기 패리티 비트수 보다 작을 경우

만큼 천공을 시켜서 전송한다. 상기

는 실제 사용되는 패리티의 길이로서, 입력되는 정보어의 길이

와 전송에 필요한 부호율로부터 구할 수 있다. 상기 단축 및 천공에 대한 상세한 설명은 생략한다.Number of bits of information word given for LDPC code

And the number of codeword bits

Encoding and decoding are performed on the. Number of parity bits

Is

to be. The number of bits of information words input to the encoder

The number of information word bits given above

If is large

Send by shortening as much. Also, the required number of parity bits

Is less than the parity bit number

Perforate as much as you send. remind

Is the length of parity actually used, and the length of information word inputted

And the code rate required for transmission. Detailed descriptions of the above shortening and drilling are omitted.

In the case of transmitting the coded data, a very good performance can be guaranteed in an additive white gauge noise (AWGN) channel, but a sufficient diversity may not be obtained in a fading channel. Therefore, we need a way to overcome this. Hereinafter, a detailed description of the present invention will be described based on the Digital Video Broadcasting the 2nd Generation Terrestrial (DVB-T2) system and the Digital Video Broadcasting Next Genration Handheld (DVB-NGH) system. However, it is obvious that the present invention is not limited to this system. In addition, the present invention will be described using a method of transmitting signaling information, but the present invention is not limited to the signaling information.

4 is a diagram illustrating an OFDM frame structure of a DVB-T2 / NGH system. 4 is composed of a plurality of frames, it can be seen that one frame is composed of a plurality of OFDM symbols (symbol). One frame includes a P1 / P2 symbol part and a data part. Signaling information is mapped and transmitted to the P1 / P2 symbol part, and data other than the signaling information is mapped to the data part and transmitted in a plurality of OFDM symbols.

As such, since the data signals are transmitted in a plurality of frames, sufficient diversity gain can be obtained, but in the case of signaling information, sufficient diversity gain can not be obtained, resulting in performance degradation.

An object of the present invention is to provide a method and apparatus for transmitting data to sufficiently obtain diversity.

Another object of the present invention is to provide a method and apparatus for generating a parity codeword group to sufficiently obtain diversity.

According to an embodiment of the present invention, in a method of transmitting data in a communication system, a process of transmitting an information word constituting a codeword in a k + s-th frame, and based on parity bits obtained by encoding the information word generating the s groups and dividing the s groups into s frames preceding the k + s-th frames, respectively.

In addition, according to an embodiment of the present invention, in the method for transmitting data in a communication system, except for the punctured bits according to a predetermined puncturing pattern of the information word constituting the codeword and the parity bits obtained by encoding the information word Transmitting the remaining bits in the k + s first frame, generating the parity bits in s groups, and transmitting the s groups in s frames preceding the k + s th frames, respectively. It includes.

According to an embodiment of the present invention, an apparatus for transmitting data in a communication system, comprising: an encoder for encoding an information word according to a predetermined method, and a puncturer for puncturing a code word encoded by the encoder according to a predetermined puncturing pattern. A parity group generator configured to preferentially configure parity bits punctured by the puncturer among s codewords output from the coder into s groups, and an information word constituting the codeword and a group of the parity bits, respectively, in k + sth frames. And a transmitter for dividing and transmitting s frames before the k + s-th frames.

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 can obtain additional diversity gain by transmitting parity bits over multiple frames.

In addition, the present invention can increase the diversity gain by transmitting additional parity bits through a plurality of frames, thereby minimizing additional modules in the encoding and decoding process.

1 shows an example of a parity check matrix of an LDPC code of length 8.
2 is a Tanner graph of an example of a parity check matrix of an LDPC code of length 8.
3 is a diagram illustrating a parity check matrix of an LDPC code used in a DVB-S2 system.
4 shows a frame structure in the form of DVB-T2 / NGH.
FIG. 5 is a diagram showing an embodiment of a frame structure according to Rule 3 according to an embodiment of the present invention; FIG.
FIG. 6 is a diagram illustrating parities constituting Q _ldpc groups. FIG.
FIG. 7 is a diagram showing an embodiment of a frame structure according to Rule 4 according to an embodiment of the present invention. FIG.
8A to 8C show an embodiment of a frame structure according to Rule 5 in accordance with an embodiment of the present invention.
9 is a diagram according to an embodiment in which the present invention is applied to a frame structure of a DVB-T2 / NGH system.
10 is a diagram according to another embodiment in which the present invention is applied to a frame structure of a DVB-T2 / NGH system.
11 is a diagram according to another embodiment in which the present invention is applied to a frame structure of a DVB-T2 / NGH system.
12 and 14 illustrate an embodiment of selecting additional parity bits according to an embodiment of the present invention.
15 is a block diagram of a transmission and reception apparatus using the frame structure of the present invention.
16 is a block diagram of a transmitter using a frame structure according to an embodiment of the present invention.
17 is a block diagram of a receiving apparatus using a frame structure according to an embodiment of the present invention.
18 is a block diagram of a transmitter using a frame structure according to an embodiment of the present invention.
19 is a block diagram of a receiving apparatus using a frame structure according to an embodiment of the present invention.
20 is a flowchart illustrating an operation of a transmitting apparatus using a frame structure according to an embodiment of the present invention.
21 is a flowchart illustrating an operation of a receiving apparatus using a frame structure according to an embodiment of the present invention.
22 is a diagram illustrating an embodiment applied to various frame structures.
FIG. 23 illustrates an embodiment for selecting additional parity bits according to an embodiment of the present invention. FIG.

Preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. It should be noted that in the following description, only parts necessary for understanding the operation according to the present invention will be described, and descriptions of other parts will be omitted so as not to distract from the gist of the present invention.

The codewords mentioned below are composed of two parts. The first part is the information word part, which means the actual information words, and the second part is additional information obtained through the encoding process by receiving the information word as the parity part. As mentioned above, the data are punctured and shortened. We generate one codeword using the method.

If sufficient diversity gain cannot be obtained in transmitting the codeword, additional diversity gain can be obtained by transmitting data through a plurality of frames. Hereinafter, various methods of transmitting data through a plurality of frames will be described in detail using a 2nd generation terrestrial (DVB-T2) and a next generation handheld (DVB-NGH) system.

<Method 1>

,

,…,

이다. 이때 추가적인 패리티 비트들은 소정의 규칙에 의하여 그룹을 형성한다. The information words comprise additional parity bits based on the parity bits and the information word bits that can be obtained by transmitting the z-th frame and encoding the information word. The additional parity bits are sent in s frames. The additional group of parity bits transmitted in each frame are G (0), G (1),... , G (s-1) and the number of bits in each group

,

,… ,

to be. At this time, the additional parity bits form a group by a predetermined rule.

The additional parity bits refer to parity bits transmitted in a different frame from the information words, in addition to the parity bits transmitted in the same frame as the information words.

Hereinafter, a rule for forming additional parity bits will be described in detail.

First, as mentioned above, a code is generated based on a (N _lpdc , K _lpdc ) code. Then, a puncturing pattern of parity bits suitable for the code can be obtained. Detailed description of the method for obtaining the perforation pattern will be omitted since it may obscure the subject matter of the present invention.

인 경우 천공 패턴에서 순차적으로

개의 원소를 선택하여 그 값을 인덱스로 하는 패리티 비트들을 천공한다. 천공하는 순서인 천공 패턴(pattern) 집합 A={1, 5, 2, 6, 4, 0, 3, 7}이라고 가정할 경우, 천공해야 할 패리티 비트의 개수가 2이면, 상기 천공 패턴 집합 A에서 2개의 숫자를 차례로 선택하면 1과 5이므로, 패리티 비트들 p₀, p₁, p₂, p₃, p₄, p₅, p₆, p₇ 중에서 p₁와 p₅은 천공을 하고 p₀, p₂, p₃, p₄, p₆, p₇은 패리티 파트로 선택되어 전송을 한다.The puncturing pattern of the parity bits means an index order of parity bits and an index order of parity bits to be punctured. That is, the number of parity bits that need to be punctured

In the sequential pattern

Elements are punctured by parity bits whose values are indexed. Assuming a puncturing pattern set A = {1, 5, 2, 6, 4, 0, 3, 7}, which is the order of puncturing, if the number of parity bits to be punctured is 2, the puncturing pattern set A in 2 when sequentially selecting the numbers 1 and 5. because the parity bits p _0, p _1, p _2, p _3, p _4, p _5, p _6, p ₇ in p ₁ and p ₅ is a perforation, and p ₀ , p ₂ , p ₃ , p ₄ , p ₆ , p ₇ are selected as the parity part to transmit.

The rules for configuring additional parity bits for the information word and parity bits are as follows.

Rule 3

Additional parity bits may be selected from parity bits and information word bits. The order of selecting additional parity bits among the parity bits is based on the order of puncturing. That is, the order of additional parity bits is determined according to the puncturing pattern. In this case, the puncturing pattern order may be used as it is, or the reverse order may be used. When additional parity bits are selected from the information word bits, the information word bits are selected in order to ensure excellent performance.

In general, the bits concatenated in the order of the puncturing pattern are configured to have little correlation with each other. That is, even though both contiguous bits undergo deep fading and recovery is difficult, the order is determined to facilitate recovery by other bits. When the bits concatenated in the order or reverse order of the puncturing patterns are mapped to the same group and transmitted in the same frame, even if the corresponding frame all undergoes deep fading, it is easy to recover by other parity bits. Therefore, when the additional parity bits are selected in the order or the reverse order of the puncturing patterns as in Rule 3, recovery is easy.

In rule 3, in order to configure additional parity bits, an order may be determined first with the remaining bits except for the punctured bits, all the unpunctured bits may be selected, and then the punctured bits may be selected.

If it is necessary to select additional parity bits after selecting all the bits punctured in rule 3, a bit can be selected using the greedy algorithm between the information word bits and the parity bits. The greedy algorithm selects the best performing bits of the codeword bits when selecting the first additional parity bits, fixes the first selected bit as the previously selected one when selecting the second bits, and performs the most performance among the remaining bits. This is a good bit to choose. When choosing a good bit of performance, consider the following:

1) the order of the variable node corresponding to the codeword bits

2) minimum cycle of the variable node corresponding to the codeword bits

3) BER performance of codeword bits

Regarding the additional parity bit selection using the greedy algorithm, two embodiments will be described below, but it is obvious that various methods may exist.

,

,…,

라 하고, 상기 값들을 모두 합한 값을

라고 한다.The number of elements of the additional parity bit group

,

,… ,

The sum of all the above values

It is called.

&Lt; Example 1 >

When the length of the parity part is larger than the sum of the number of elements of the group of additional parity bits, the group of additional parity bits is formed according to the order of the puncturing pattern in forming the group of additional parity bits based on the parity bits of the codeword. do.

Hereinafter, rule 3 for forming a group of parity bits will be described with reference to FIG. 5.

이며, 패리티 파트의 길이는 8로

이며, 천공 순서를 의미하는 천공 패턴(pattern) 집합 A={1, 5, 2, 6, 4, 0, 3, 7} 이라고 가정한다.Referring to FIG. 5, a codeword is composed of an information word part and a parity part. If the length of the information word part after shortening is 7, the information word part

The length of the parity part is 8

It is assumed that a set of puncture patterns A = {1, 5, 2, 6, 4, 0, 3, 7}, which means a puncturing order.

, 추가적인 패리티 비트들이 전송되는 프레임의 개수 s=3, 각각의 프레임으로 전송되는 추가적인 패리티 비트들의 개수

=

=

= 2일 경우에 대하여 다음과 같이 추가적인 패리티 비트들의 그룹을 구성한다. Number of bits perforated

Number of additional parity bits transmitted s = 3, number of additional parity bits transmitted in each frame

=

=

For the case of = 2, a group of additional parity bits is configured as follows.

First, since the number of bits to be punctured is 2, p ₁ and p ₅ are punctured according to the puncturing pattern set A. FIG.

The elements constituting the group of additional parity bits are identical to the puncturing order among the bits p ₀ , p ₂ , p ₃ , p ₄ , p ₆ , p ₇ except for the punctured bits p ₁ and p ₅ . First group G (0) = {p ₂ , p ₆ }, second group G (1) = {p ₄ , p ₀ }, third group G (2) = {p ₃ , p ₇ } Becomes The first group transmits in the k + 2th frame, the second group transmits in the k + 1th frame, and the third group transmits in the kth frame.

Alternatively, when the elements constituting the group of additional parity bits are determined to be identical to the reverse order of the punctuation order among the bits except the punctured bits, the first group G (0) = {p ₇ , p ₃ }, two The first group G (1) = {p ₀ , p ₄ }, and the third group G (2) = {p ₆ , p ₂ }. The first group transmits in the k + 2th frame, the second group transmits in the k + 1th frame, and the third group transmits in the kth frame.

In the above embodiment, the determined groups are sequentially transmitted, but the order of group transmission may be changed.

<Example 2>

If the length of the parity part is less than or equal to the sum of the number of elements of the additional parity bits group, additional parity bits are configured based on the parity bits and the information word of the codeword.

이며, 패리티 파트의 길이는 8로

이며, 천공 순서를 의미하는 천공 패턴(pattern) 집합 A={1, 5, 2, 6, 4, 0, 3, 7} 이라고 가정한다.A codeword is composed of an information word part and a parity part. If the length of the information word part after shortening is 7, the information word part

The length of the parity part is 8

It is assumed that a set of puncture patterns A = {1, 5, 2, 6, 4, 0, 3, 7}, which means a puncturing order.

, 추가적인 패리티 비트들이 전송되는 프레임의 개수 s=3, 각각의 프레임으로 전송되는 추가적인 패리티 비트들의 개수

=

=

= 3일 경우에 대하여 다음과 같이 추가적인 패리티 비트들을 구성한다. Number of bits perforated

Number of additional parity bits transmitted s = 3, number of additional parity bits transmitted in each frame

=

=

For case of = 3, additional parity bits are configured as follows.

First, since the number of bits to be punctured is 2, p ₁ and p ₅ are punctured according to the puncturing pattern set A. FIG.

The elements that make up the group of additional parity bits are determined in the same order of puncturing, so that the first group G (0) = {p ₂ , p ₆ , p ₄ }, the second group G (1) = {p ₀ , p ₃ , p ₇ }, the third group G (2) = {p ₁ , p ₅ , i ₀ }. In this case, it can be seen that one bit i ₀ is selected from the parity bits p ₅ and p ₁ and the codeword bits that were punctured in the third group. The first group transmits in the k + 2th frame, the second group transmits in the k + 1th frame, and the third group transmits in the kth frame.

In addition, when the elements constituting the group of additional parity bits are determined in the reverse order of puncturing, the first group G (0) = {p ₇ , p ₃ , p ₀ } and the second group G (1) = {p ₄ , p ₆ , p ₂ }, and the third group G (2) = {p ₅ , p ₁ , i ₀ }. Even in this case, it can be seen that one bit i ₀ is selected from the parity bits p ₅ and p ₁ and the codeword bits that were punctured in the third group. The first group transmits in the k + 2th frame, the second group transmits in the k + 1th frame, and the third group transmits in the kth frame.

In the above embodiment, the determined groups are sequentially transmitted, but the order of group transmission may be changed.

Hereinafter, a method of generating a group of a plurality of additional parity bits according to rule 3 based on the information word bits and the parity bits will be described in detail.

FIG. The length of the LDPC codeword is mentioned in 3 N _ldpc an example a case in which the length of an information word is K _ldpc example, the parity bits are {p _0, p _1, ... , p _{N ldpc- K ldpc -1} } and can configure a group of Q _ldpc parity bits.

6 illustrates parity bits constituting Q _ldpc groups.

An index group representing an index value of parity bits constituting the j-th group among the groups may be expressed as in Equation 3 below.

Based on Equation 3, an index group of parity bits for a puncturing pattern meaning a puncturing order may be obtained as in Equation 4 below.

번째 원소는

인덱스 그룹의 i번째 원소와 같다. 상기

는 하기의 표 1과 같다.That is, the puncturing pattern set A

The first element

Same as the i th element of the index group. remind

Is shown in Table 1 below.

Naturally, Table 1 may be changed.

의 원소의 개수를

라고 하면 그룹

(

)의 패리티 비트의 인덱스 그룹은 다음 수학식 5와 같이 구할 수 있다. The number of bits to be punched is called N _punc , and the group

The number of elements in

Speaking of groups

(

The index group of the parity bit of) can be obtained as shown in Equation 5 below.

번째 추가적인 패티리 비트들의 그룹을 구성하는 패리티 비트의 인덱스는 상기 천공 순서를 의미하는 천공 패턴 집합 A에서 천공이 되는 패리티 비트에 상응하는 인덱스들을 제외하고, 순차적으로

개를 선택한 것을 의미한다. In other words,

The indexes of the parity bits constituting the first additional group of parity bits are sequentially except for the indexes corresponding to the parity bits to be punctured in the puncturing pattern set A representing the puncturing order.

Means choosing a dog.

의 패리티 비트의 인덱스 그룹을 구할 수 있다.Alternatively, the index of the parity bits constituting the group of additional parity bits may be selected in the reverse order of the puncturing pattern, in which case, the following equation (6)

An index group of parity bits of may be obtained.

는 패리티 비트들의 개수이다. 즉,

번째 추가적인 패티리 비트들의 그룹을 구성하는 패리티 비트의 인덱스는 상기 천공 순서를 의미하는 패턴 집합 A에서 마지막 비트부터 역순으로

개를 선택한 것을 의미한다.In Equation 6

Is the number of parity bits. In other words,

The index of the parity bits constituting the first additional group of parity bits is reversed from the last bit in the pattern set A representing the puncturing order.

Means choosing a dog.

An element of a group of parity bits transmitted in a plurality of frames based on a group of indices obtained using Equation 5 or Equation 6 may be expressed as Equation 6 below.

번째 인덱스들의 그룹

에 속한 값들을 인덱스로 하는 패리티 비트들이

번째 추가적인 패리티 비트들의 그룹

을 구성함을 의미한다.In Equation 7, p _k denotes a k th parity bit of the parity check code. Therefore, Equation 7 is obtained from Equation 6 above.

Of the first index

Parity bits that index the values in

Group of the first additional parity bits

It means to configure.

Therefore, Equation 7 indicates that the group is generated to satisfy the above-described rule 3.

Rule 4

In configuring additional parity based on the parity bits and the information word, the parity bits are sequentially selected to select groups G (0), G (1),... , And then maps to G (s-1). When selecting information words, information words are selected in order to ensure excellent performance.

In configuring the additional parity bits according to the rule 4, it is possible to determine the order of configuring the additional parity bits among the remaining bits except the bits that are first punctured, and to select the punctured bits after selecting all the unpunctured bits. Can be.

If additional parity bits are to be selected after selecting all the bits punctured in rule 4, bits that can guarantee excellent performance between the information word bits and the parity bits can be selected using the greedy algorithm. The greedy algorithm selects the best performing bits of the codeword bits when selecting the first additional parity bits, and fixes the first selected bit as the previously selected one when selecting the second bits, and selects the most among the remaining bits. The best bit is to choose a good performance. When choosing a good bit of performance, consider the following:

1) the order of the variable node corresponding to the codeword bits

2) minimum cycle of the variable node corresponding to the codeword bits

3) BER performance of codeword bits

In the parity check matrix, when the parity part has a double diagonal structure, contiguous bits are connected to the same check node, and therefore, an association is formed between the contiguous bits and it is not preferable to transmit them on the same channel. Therefore, you will experience different channels by mapping the parity bits to different groups sequentially.

<Example 3>

7 illustrates an embodiment of a frame structure according to rule 4. FIG.

이며, 패리티 파트의 길이는 8로

이며, 천공 순서를 의미하는 천공 패턴(pattern) 집합 A={1, 5, 2, 6, 4, 0, 3, 7} 이라고 가정한다.Referring to FIG. 7, a codeword is composed of an information word part and a parity part. If the length of the information word part after shortening is 7, the information word part

The length of the parity part is 8

It is assumed that a set of puncture patterns A = {1, 5, 2, 6, 4, 0, 3, 7}, which means a puncturing order.

, 추가적인 패리티 비트들이 전송되는 프레임의 개수 s=3, 각각의 프레임으로 전송되는 비트들의 개수

일 경우에 대하여 다음과 같이 추가적인 패리티 비트들을 구성한다. 이때, 패리티 파트에 해당하는 패리티 검사 행렬의 구조는 이중 대각 행렬이라고 가정한다. Number of bits perforated

, Number of frames in which additional parity bits are transmitted, s = 3, number of bits transmitted in each frame

For one case, additional parity bits are configured as follows. In this case, it is assumed that the structure of the parity check matrix corresponding to the parity part is a double diagonal matrix.

Since the number of bits to be punctured is 2, p ₁ and p ₅ are punctured according to the puncturing pattern set A.

Rule 4 maps each bit to each group sequentially, so that p ₀ is the first group, and p ₂ is the first of p ₀ , p ₂ , p ₃ , p ₄ , p ₆ , p ₇ , except for the punctured parity bits. in the second group, the p ₃ is, p ₄ is, p ₇ are mapped to the third group in the first group, p ₆ is the second group to the third group. Thus, the first group G (0) = {p ₀ , p ₄ }, the second group G (1) = {p ₂ , p ₆ }, and the third group G (2) = {p ₃ , p ₇ } . The first group transmits on the k + 2 th frame, the second group transmits on the k + 1 th frame, and the third group transmits on the k th frame. In the above example, it is described that the determined groups are transmitted in order, but the order of group transmission may be changed.

<Example 4>

이며, 패리티 파트의 길이는 8로

이며, 천공 순서를 의미하는 천공 패턴(pattern) 집합 A={1, 5, 2, 6, 4, 0, 3, 7} 이라고 가정한다.If the length of the information word part after shortening is 7, the information word part

The length of the parity part is 8

It is assumed that a set of puncture patterns A = {1, 5, 2, 6, 4, 0, 3, 7}, which means a puncturing order.

, 추가적인 패리티 비트들이 전송되는 프레임의 개수 s=3, 각각의 프레임으로 전송되는 비트들의 개수

일 경우에 대하여 다음과 같이 추가적인 패리티 비트들의 그룹을 구성한다. 이때, 패리티 파트에 해당하는 패리티 검사 행렬의 구조는 이중 대각 행렬이라고 가정한다.
Number of bits perforated

, Number of frames in which additional parity bits are transmitted, s = 3, number of bits transmitted in each frame

For one case, a group of additional parity bits is configured as follows. In this case, it is assumed that the structure of the parity check matrix corresponding to the parity part is a double diagonal matrix.

Since the number of bits to be punctured is 2, p ₁ and p ₅ are punctured according to the puncturing pattern set A.

는 상기 패리티 비트들의 개수 8보다 크므로, 마지막 그룹을 구성하는 비트들을 선택할 때는 정보어를 선택한다. 즉, 첫 번째 그룹 G(0)={p₀, p₄, p₁}, 두 번째 그룹 G(1)={p₂, p₆, p₅}, 세 번째 그룹 G(2)={p₃, p₇, i₀}이 된다. The number of additional parity bits forming a group

Since P is greater than 8 of the parity bits, an information word is selected when selecting bits forming the last group. That is, the first group G (0) = {p ₀ , p ₄ , p ₁ }, the second group G (1) = {p ₂ , p ₆ , p ₅ }, the third group G (2) = {p ₃ , p ₇ , i ₀ }

The first group transmits on the k + 2 th frame, the second group transmits on the k + 1 th frame, and the third group transmits on the k th frame. In the above embodiment, it is described that the determined groups are transmitted in order, but the order of group transmission may be changed.

Hereinafter, a method of generating additional parity bits using a method other than the method described above will be described.

Diversity gain can be obtained while lowering the actual code rate by transmitting unperforated parity bits and codewords in the same frame and transmitting additional parity bits in another frame to obtain additional diversity gain.

<Method 2>

Parity bits that are not punctured are transmitted in the z-th frame, which is the same frame as the codeword, and additional parity bits obtained from the information word bits and the parity bits are grouped according to a predetermined rule, group G (0), G (1),... It consists of G (s-1) and transmits s frames.

The additional parity bits mean parity bits transmitted in a previous frame without being transmitted in the same frame as the information word, in addition to parity bits transmitted in the same frame as the information word. The G (0), G (1),... , G (s-1) means a group of bits consisting of additional parity bits.

Rule 5

The additional parity bits are selected from the parity bits and the information word bits, and the parity bits that are not transmitted in the same frame as the information word are preferentially selected. Parity bits that are not transmitted in the same frame as the information word mean punctured parity bits. When the additional parity bits are selected from the parity bits, the order of the additional parity bits constituting each group is determined according to the order of the puncturing patterns. In this case, the puncturing pattern order may be used as it is, or the reverse order may be used. When selecting information words, information words are selected in order to ensure excellent performance.

In configuring the additional parity bits according to rule 5, first, parity bits that are not transmitted in a frame in which an information word is transmitted may be selected as additional parity bits, and after selecting all parity bits that are not transmitted in the same frame as the information word, From among codeword bits consisting of information bits and parity bits, a bit to configure additional parity bits is selected.

If additional parity bits are to be selected after selecting all the parity bits punctured in Rule 5, a bit can be selected using the greedy algorithm between the information word bits and the parity bits. The greedy algorithm selects the best performing bits of the codeword bits when selecting the first additional parity bits, fixes the first selected bit as the previously selected one when selecting the second bits, and performs the most performance among the remaining bits. This is an excellent bit to choose. When choosing a good bit of performance, consider the following:

1) the order of the variable node corresponding to the codeword bits

2) minimum cycle of the variable node corresponding to the codeword bits

3) BER performance of codeword bits

Example 5

8A to 8C show an embodiment of a frame structure according to rule 5. FIG.

이며, 패리티 파트의 길이는 8로

이며, 천공 순서를 의미하는 천공 패턴(pattern) 집합 A={1, 5, 2, 6, 4, 0, 3, 7} 이라고 가정한다.Referring to FIG. 8A, a codeword is composed of an information word part and a parity part. If the length of the information word part after shortening is 7, the information word part

The length of the parity part is 8

It is assumed that a set of puncture patterns A = {1, 5, 2, 6, 4, 0, 3, 7}, which means a puncturing order.

=3, 추가적인 패리티 비트들이 전송되는 프레임의 개수 s=3, 각각의 프레임으로 전송되는 추가적인 패리티 비트들의 개수

일 경우에 대하여 다음과 같이 추가적인 패리티 비트들을 구성한다. If the number of parity bits transmitted in the same frame as the information word is 5, the number of parity bits to be punctured

= 3, the number of additional parity bits transmitted in the frame s = 3, the number of additional parity bits transmitted in each frame

For one case, additional parity bits are configured as follows.

=3이므로

는 천공되며, 정보어와 동일 프레임에 전송되는 패리티 비트들은

이다. remind

= 3

Is punctured, and the parity bits transmitted in the same frame as the information word are

to be.

를 선택하여 G(0)={p₁, p₅}를 구성한다. 두 번째 그룹은 천공 패턴의 세 번째 원소인 2를 인덱스로 하는 패리티 비트

를 선택한다. 천공되는 패리티 비트들을 모두 선택하였으므로 정보어 내지 천공되지 않는 패리티 비트들 중에 하나를 선택하여 두 번째 그룹을 구성한다. 만약 첫 번째 정보어 비트를 선택하였다고 하면, G(1)={p₂, i₀}이다. 세 번째 그룹을 구성하기 위하여 정보어 내지 천공되지 않는 패리티 비트들 중에서 선택하도록 하며

를 선택하였다고 하면, 세 번째 그룹 G(2)={i_1, i₂}가 된다. 첫 번째 그룹은 k+2 번째 프레임에 전송하며, 두 번째 그룹은 k+1 번째 프레임에 전송하며, 세 번째 그룹은 k 번째 프레임에 전송한다. Since each bit is mapped to each group in puncture order according to rule 5, the first group is parity bits whose index values are the first and second elements of the puncturing pattern, 1 and 5

Select to configure G (0) = {p ₁ , p ₅ }. The second group is a parity bit with an index of 2, the third element of the perforation pattern

Select. Since all the parity bits are punctured, one of the information words or the non-punctured parity bits is selected to form a second group. If the first information word bit is selected, G (1) = {p ₂ , i ₀ }. In order to form a third group, one of the information words or the non-perforated parity bits is selected.

If is selected, the third group G (2) = {i _1, i ₂ }. The first group transmits on the k + 2 th frame, the second group transmits on the k + 1 th frame, and the third group transmits on the k th frame.

Referring to FIG. 8B, when each bit is mapped to each group in the reverse order of the puncturing pattern, if the _first bit is selected from the last bit among the parity bits {p ₁ , p ₅ , p ₂ } that are punctured among the elements of the puncturing pattern set A, FIG. The first group G (0) = {p ₂ , p ₅ }, the second group G (1) = {p ₁ , i ₀ }, and the third group G (2) = {i _1, i ₂ }. The first group transmits on the k + 2 th frame, the second group transmits on the k + 1 th frame, and the third group transmits on the k th frame.

로 나열하고, 정보어와 동일 프레임으로 전송하는 패리티 비트들은 상기 나열된 패리티 비트들중에서 5개 비트들을 순차적으로 선택하므로

이고, G(0)은

를 선택한다. 또한 G(1)은

을 선택하고 정보어중에서 한 비트를 선택한다. 또한 G(2)는 그리디 알고리즘에 따라 정보어중에서 두 비트를 선택한다.Referring to FIG. 8C, in order to facilitate selection of parity bits transmitted in each frame, the parity bits are sequentially transmitted after permutation of the parity bits in the reverse order of the puncturing pattern. That is, since the puncture pattern is A = {1, 5, 2, 6, 4, 0, 3, 7}, the parity bits are in the reverse order of the puncture pattern.

The parity bits transmitted in the same frame as the information word are selected sequentially from five bits among the parity bits listed above.

And G (0) is

Select. G (1) is also

Select and select one bit from the information word. G (2) also selects two bits from the information word according to the greedy algorithm.

In the above embodiment, it is described that the determined groups are transmitted in order, but the order of group transmission may be changed.

<Example 6>

이며, 패리티 파트의 길이는 8로

이며, 천공 순서를 의미하는 천공 패턴(pattern) 집합 A={1, 5, 2, 6, 4, 0, 3, 7} 이라고 가정한다.If the length of the information word part after shortening is 7, the information word part

The length of the parity part is 8

It is assumed that a set of puncture patterns A = {1, 5, 2, 6, 4, 0, 3, 7}, which means a puncturing order.

=3, 추가적인 패리티 비트들이 전송되는 프레임의 개수 s=3, 각각의 프레임으로 전송되는 비트들의 개수

일 경우에 대하여 다음과 같이 추가적인 패리티 비트들을 구성한다. Number of bits perforated

= 3, the number of frames in which additional parity bits are transmitted s = 3, the number of bits in each frame

For one case, additional parity bits are configured as follows.

=3이므로

는 천공되며, 정보어와 동일 프레임에 전송되는 패리티 비트들은

이다. remind

= 3

Is punctured, and the parity bits transmitted in the same frame as the information word are

to be.

이며, 이중

가 첫 번째 그룹 G(0)으로 전송되었으므로 세 번째 그룹 G(2)에서는

를 전송한다.According to rule 5, if the remaining parity bits except for the parity bits transmitted in the same frame as the information word are mapped to each group in puncture order, the first group G (0) = {p ₁ , p ₅ , p ₂ , i ₀ } , The second group G (1) = {i _1, i _2, i ₃ i ₄ }, and the third group G (2) = {i _5, i ₆ , p ₆ , p ₄ }. When configuring the third group G (2), list the parity bits according to the puncturing pattern

Is, double

Was sent to the first group G (0), so in the third group G (2)

Send it.

이며, 이중

이 첫 번째 그룹 G(0)으로 전송되었으므로 세 번째 그룹 G(2)에서는 패리티 비트들을 천공 패턴의 역순으로 나열하였을 경우 첫 번째와 두 번째 패리티 비트인

를 전송한다.For each of the parity bits except for the parity bits P, the first group G (0) = {p ₂ , p ₅ , p ₁ , i ₀ }, and the second group are mapped to each group in the reverse order of puncturing. Group G (1) = {i _1, i _2, i ₃ , i ₄ }, and the third group G (2) = {i _5, i ₆ , p ₇ , p ₃ }. When configuring the third group G (2), if you list the parity bits in the reverse order of the puncturing pattern,

Is, double

Is transmitted to the first group G (0), so when the third group G (2) lists the parity bits in the reverse order of the puncturing pattern, the first and second parity bits

Send it.

The first group transmits on the k + 2 th frame, the second group transmits on the k + 1 th frame, and the third group transmits on the k th frame.

In the above embodiment, it is described that the determined groups are transmitted in order, but the order of group transmission may be changed. In addition, the method of selecting the information word bits in the above embodiment is a matter of course.

Hereinafter, a method of configuring additional parity bits when encoding and decoding based on a parity check matrix used in DVB-S2, DVB-T2, DVB-C2, and DVB-NGH systems will be described.

= N_lpdc - K_lpdc = 9000이다. 상기에서 설명한 바와 같이 LDPC 부호의 부호어 파트는 360단위로 블록을 구성하고 있다. 패리티 검사 행렬에서 정보어에 해당하는 7200개의 열을 360개씩 그룹화(grouping)하여, 총 20개의 열 그룹(column group)을 생성한다. 또한 패리티 파트는 상기 수학식 2와 같이 360개의 원소로 구성되어 있는 패리티들의 인덱스들을 나타내는 Q_ldpc개의 그룹을 표현할 수 있다. 부호율 4/9일 경우 Q_ldpc = 25이다. 또한 천공 패턴은 상기 표 1 중에 하나를 사용할 수 있다. First, the code rate of the parity check matrix is 4/9. The number of bits of information word K _lpdc = 7200, the number of bits of codeword N _lpdc = 16200, the number of parity bits.

= N _lpdc -K _lpdc = 9000. As described above, the codeword part of the LDPC code forms a block in 360 units. A total of 20 column groups are generated by grouping 7200 columns corresponding to information words in the parity check matrix by 360. In addition, the parity part may express Q _ldpc groups representing indices of parities composed of 360 elements as shown in Equation 2 above. For code rate 4/9, Q _ldpc = 25. In addition, the perforation pattern may use one of Table 1 above.

For convenience of explanation, the set of indices constituting the column group of the information word part may be expressed as in Equation 8.

In Equation 8, the symbol [a] means an integer smaller than a. For example, [0.2] = 0 and [11.8] = 11. In addition, N _info refers to the number of column groups of the information word part, and when N = R, _4/9 , N _info = 20. In addition, K _bch means the number of information word bits of the BCH when the LDPC code and the BCH code are concatenated.

Hereinafter, a procedure of configuring optimal additional parity bits according to length will be described. In this case, it is assumed that the additional parity bits are selected in the reverse order of puncturing.

When the number of input codeword bits is 7200, an order of generating additional parity bits is as follows.

Since a puncturing and a shortening are not necessary when the number of codeword bits to be input is 7200, first, additional parity bits are selected in the reverse order of puncturing. That is, the index value is selected in the following order to select information word bits and parity bits corresponding to the index value. Hereinafter, I (a) means a set of indices of parity bits as defined in Equation 3, and X (a) means a set of indices of information word bits as defined in Equation 8. Therefore, in the following, I (a) means that parity bits should be selected to index elements of I (a), and X (a) should select information word bits to index elements of X (a). Means that.

I (11) → I (7) → I (19) → I (21) → I (0) → I (14) → I (1) → I (23) → I (16) → I (3) → I (12) → I (22) → I (10) → I (24) → I (2) → I (17) → I (5) → I (20) → I (15) → I (8) → I (13) → I (9) → I (18) → I (4) → I (6)

 I (9) → I (26) → I (3) → I (15) → I (30) → I (13) → I (6) → I (19) → I (34) → I (16) → I (1) → I (23) → I (4) → I (17) → I (22) → I (24) → I (7) → I (11) → I (31) → I (10) → I (8) → I (2) → I (35) → I (28) → I (20) → I (18) → I (25) → I (33) → I (0) → I (21) → I (12) → I (5) → I (32) → I (29) → I (14) → I (27)

If additional parity bits other than the parity bits are needed, they are selected in the following order.

I (11) → I (7) → I (19) → I (21) → I (0) → I (14) → I (1) → I (23) → I (16) → I (3) → I (12) → I (22) → I (10) → I (24) → I (2) → I (17) → I (5) → I (20) → I (15) → I (8) → I (13) → I (9) → I (18) → I (4) → I (6) → X (5) → X (6) → X (7) → X (8) → X (9) → X (10) → X (11) → X (12) → X (13) → X (14) → X (15) → X (16) → X (17) → X (18) → X (19) → X (0) → X (1) → X (2) → X (3) → X (4)

X (4) → X (5) → I (9) → I (26) → X (6) → X (7) → X (8) → I (3) → I (15) → I (30) → I (13) → I (6) → I (19) → I (34) → I (16) → I (1) → I (23) → I (4) → I (17) → I (22) → I (24) → I (7) → I (11) → I (31) → I (10) → I (8) → I (2) → I (35) → I (28) → I (20) → I (18) → I (25) → I (33) → I (0) → I (21) → I (12) → I (5) → I (32) → I (29) → I (14) → I (27)

If additional parity bits are needed later, the transmission is repeated in the same order as the above transmission.

When the number of input codeword bits is 3600, the order of generating additional parity bits is as follows.

First, when the number of codeword bits input is 3600, 10 information word blocks are shortened and 12 parity blocks are punctured. Therefore, the number of blocks that are not punctured is 25-12 = 13 blocks. First, the blocks that are not punctured are selected in the following order.

I (11) → I (7) → I (19) → I (21) → I (0) → I (14) → I (1) → I (23) → I (16) → I (3) → I (12) → I (22) → I (10)

I (9) → I (26) → I (3) → I (15) → I (30) → I (13) → I (6) → I (19) → I (34) → I (16) → I (1) → I (23) → I (4) → I (17) → I (22) → I (24) → I (7) → I (11) → I (31) → I (10) → I (8) → I (2) → I (35) → I (28) → I (20) → I (18) → I (25) → I (33) → I (0)

If additional parity bits other than the parity bits are required, they are selected in the following order.

X (5) → X (6) → X (7) → X (8) → X (9) → X (19) → X (0) → X (1) → I (11) → I (7) → X (2) → X (3) → I (19) → I (21) → I (0) → I (14) → I (1) → I (23) → I (16) → I (3) → I (12) → I (22) → I (10) → I (24) → I (2) → I (17) → I (5) → I (20) → I (15) → I (8) → I (13) → I (9) → I (18) → I (4) → I (6)

I (21) → I (12) → I (5) → I (32) → I (29) → I (14) → I (27) → X (4 → X (5 → X (6) → I ( 9) → X (8) → I (26) → I (3) → I (15) → I (30) → I (13) → I (6) → I (19) → X (4) → I ( 34) → I (16) → I (1) → I (23) → I (4) → I (17) → I (22) → I (24) → I (7) → X (5) I (11 ) → I (31) → I (10) → I (8) → X (6) → I (2) → I (35) → I (28) → I (20) → I (18) → I (25 ) → I (33) → I (0) → I (21) → I (12) → I (5) → I (32) → I (29) → I (14) → I (27)

If additional parity bits are needed later, the codeword bits are repeatedly transmitted.

As can be seen from above, the order of selecting additional parity bits having optimal performance may vary according to the length of the information word input. However, in order to simplify the configuration of the system, the selection in the order described below can optimize the efficiency of the system while similar to the result having the best performance.

Method 1 does not transmit the parity bits in the same frame as the information word, so when additional parity bits are selected, they are selected in the following order.

1) Select non-perforated parity bits in the order or reverse order of the puncturing pattern.

2) If additional parity bits are needed, select the parity bits that were punctured in the order of the puncturing pattern or the reverse order.

3) If additional parity bits are needed further, select from order 3 information word bits.

4) If more parity bits are needed, select the parity bits in the order or reverse order of the puncturing pattern.

5) If more parity bits are needed, select information word bits of order 12.

6) If additional parity bits are needed, repeat from step 1).

In addition to the above-described order, it may be selected in the following order.

1) Select non-perforated parity bits in the order or reverse order of the puncturing pattern.

2) If additional parity bits are needed, select the parity bits that were punctured in the order of the puncturing pattern or the reverse order.

3) If additional parity bits are needed further, select from order 3 information word bits.

4) If more parity bits are needed, select information word bits of order 12.

5) If more parity bits are needed, select the parity bits in the order or reverse order of the puncturing pattern.

6) If additional parity bits are needed, repeat from step 1).

In addition to the above-described order, it may be selected in the following order.

1) Select non-perforated parity bits in the order or reverse order of the puncturing pattern.

2) If additional parity bits are needed, select the parity bits that were punctured in the order of the puncturing pattern or the reverse order.

3) If additional parity bits are needed, select the parity bits in the order or reverse order of the puncturing pattern.

4) If additional parity bits are needed, select from the order 12 information word bits.

5) If more parity bits are needed, select information word bits of order two.

6) If additional parity bits are needed, repeat from step 1).

Method 2 transmits the parity bits in the same frame as the information word, so the unperforated parity bits are sent in the same frame as the information word. In addition, when additional parity bits are selected, they are selected in the following order.

1) If additional parity bits are needed, select the parity bits that were punctured in the order of the puncturing pattern or the reverse order.

2) If additional parity bits are needed further, select from order 3 information word bits.

3) If additional parity bits are needed, select the parity bits in the order or reverse order of the puncturing pattern.

4) If more parity bits are needed, select information word bits of order 12.

5) If additional parity bits are needed, select non-perforated parity bits in the order or reverse order of the puncturing pattern.

6) If additional parity bits are needed, repeat from step 1).

In addition to the above-described order, it may be selected in the following order.

1) If additional parity bits are needed, select the parity bits that were punctured in the order of the puncturing pattern or the reverse order.

2) If additional parity bits are needed further, select from order 3 information word bits.

3) If more parity bits are needed, select information word bits of order 12.

4) If additional parity bits are needed, select the parity bits to be punctured and the parity bits not to be punctured in the order or reverse order of the puncturing pattern.

5) If more parity bits are needed, repeat from step 1).

In addition to the above-described order, it may be selected in the following order.

1) If additional parity bits are needed, select the parity bits that were punctured in the order of the puncturing pattern or the reverse order.

2) If additional parity bits are needed, select the parity bits in the order or reverse order of the puncturing pattern.

3) If additional parity bits are needed further, select from order 3 information word bits.

4) If more parity bits are needed, select information word bits of order 12.

5) If more parity bits are needed, repeat from step 1).

In the above description, the LDPC codes used in the DVB-S2 to DVB-T2, DVB-C2, and DVB-NGH have been described based on codes having a code rate of 1/2, but the same method can be applied to other code rates.

Hereinafter, an embodiment of the present invention will be described based on a frame structure configured for transmitting signaling bits in DVB-T2 / NGH.

9 is a diagram illustrating a method for effectively transmitting signaling information. Hereinafter, a method for obtaining additional diversity effects when transmitting signaling bits will be described in detail with reference to FIG. 9.

 The signaling information word used in the DVB-T2 / NGH system includes a configure information part and a dynamic information part.

The configure information part of the information parts means information that does not change for several frames. The dynamic information part is composed of information that changes every frame. Since the configure information part and the dynamic information part include information on data of the current frame, it is very important to receive the error information without error. If an error occurs as a result of the CRC test, it may be determined that data information of the corresponding frame cannot be received.

Since the same information is received for several frames, the configure information part has the capability of correcting an error by obtaining a diversity gain using the stored information. On the other hand, since the dynamic information part is not repeatedly transmitted as information that changes every frame, and is only transmitted for a short time interval as a P2 symbol, sufficient diversity gain may not be obtained, and error correction capability may be significantly degraded.

로 하는 다수 그룹 G(0), G(1), …, G(s-1)으로 분리하고 s개의 프레임에 전송한다. 상기 도 9에서는 s=3인 경우에 한하여 도시하고 있지만 s는 임의로 결정할 수 있음은 당연하다. 각 그룹에 속한 비트들은 동일 프레임에 전송한다. 상기에서 패리티 부호어 비트들을 다수의 그룹으로 분리하는 방법은 상술한 규칙 3과 규칙 4를 적용할 수 있다. 또한 상기 예에서는 G(0), G(1), …, G(s-1) 이 순차적으로 프레임에 매핑되어 있으나, 그룹을 프레임에 매핑하는 방법은 다양하게 변경할 수 있음은 당연하다.Therefore, as illustrated in FIG. 9, the above-described method 1 is applied only to the dynamic information part to transmit additional parity bits in a plurality of frames. That is, after encoding the dynamic information part by a predetermined method, additional parity bits are generated, and the additional parity bits are determined as the number of elements of the group.

Multiple groups G (0), G (1),... , Separated by G (s-1) and transmitted in s frames. In FIG. 9, only s = 3, but s can be arbitrarily determined. The bits belonging to each group are transmitted in the same frame. The above-described rule 3 and rule 4 may be applied to a method of dividing the parity codeword bits into a plurality of groups. In the above example, G (0), G (1),... Although G (s-1) is sequentially mapped to a frame, the method of mapping a group to a frame may be variously changed.

As mentioned above, it is possible to know whether an error has occurred in the signaling information through a CRC check. Therefore, if the CRC check determines that an error has occurred, G (0), G (1),... , G (s-1) is input to perform additional decoding on the dynamic information. After performing the decoding operation, CRC check for signaling can be performed once again to determine whether an error has occurred.

If an error occurs in additional parity bits in the signaling information parts, a CRC check error may continue even though dynamic information is restored through the additional decoding. Therefore, to solve this problem, additional CRC is used for dynamic information.

10 is a diagram illustrating a case where an additional CRC is used for the dynamic information.

If the dynamic information is transmitted in the k + 3 th frame, the CRC for the dynamic information may be transmitted in one of the frames in which an additional parity bit is transmitted. In FIG. 10, it is shown that the k + 2th frame is transmitted. In the above method, CRC may or may not be included in an information word when dynamic information is encoded.

In the DVB-T2 / NGH system, encoding information includes configurable information, dynamic information, extension part, CRC, and padding bits as information information. In the present invention, to obtain a diversity gain by performing additional encoding on the dynamic information among the information words.

Assuming that the method proposed by the present invention is used, when transmitting as shown in FIG. 9 and FIG. 10, the transmitter should perform additional encoding on the dynamic information, and when the CRC error of the signaling information occurs in the receiver, An additional decoding process should be performed. In addition, when the puncturing and shortening patterns for encoding the dynamic information are different from the puncturing and shortening patterns for the signaling, an additional module may be needed for encoding and decoding the dynamic information.

9 and 10 need to perform encoding and decoding twice for signaling information. However, by using the above-mentioned method 2, encoding can be performed only once, and additional puncturing and shortening patterns are unnecessary. In the decoding process, additional modules can be minimized.

11 shows an embodiment in which Method 2 is applied to a DVB-T2 / NGH system.

LDPC encoding is performed by receiving the configurable information and dynamic information. In general, since signaling is variable, shortening and puncturing are performed to adaptively encode a variable length.

In the NGH system, since configurable information and dynamic information are encoded, the L1 signaling of FIG. 11 refers to all kinds of L1 signaling information that is not limited to configurable and dynamic.

인 추가적인 패리티 비트들의 그룹 G(0), G(1), G(2)는 k+2 번째 프레임, k+1 번째 프레임, k 번째 프레임으로 각각 전송된다.Parity bits other than the punctured bits are transmitted in a k + 3 th frame, which is the same frame as the signaling bits of FIG. 11. The number of elements determined by Rule 5

Groups of additional parity bits, G (0), G (1), and G (2), are transmitted in k + 2 th frames, k + 1 th frames, and k th frames, respectively.

In FIG. 11, it is assumed that the number of frames in which additional parity bits are transmitted is three, but as mentioned above, this is variable. In addition, the location of transmitting the group of additional parity bits G (0), G (1), G (2) can also be variously changed. That is, it may be transmitted in a P1 / P2 symbol or may be transmitted in an auxiliary stream. It can also be transmitted using a general data stream.

In addition, as mentioned above, it is not limited to being transmitted in the k + 2 th frame, the k + 1 th frame, the k th frame, but may be considered to be transmitted in s previous frames.

22 illustrates a method for transmitting additional parity bits in s previous frames.

FIG. 22A illustrates that additional parity bits are transmitted to several frames before the concatenated previous frame. In the DVB-NGH system, NGH frames are transmitted using a Future Extension Frame (FEF) present in the DVB-T2. Therefore, the additional parity bits of the present invention are not limited to transmitting in s contiguous frames, but may be applied by transmitting in s frames before a frame for transmitting an information word.

FIG. 22B illustrates that additional parity bits may be applied to other RF bands. That is, the additional parity bits of the present invention are not limited to frames transmitted on the same RF channel.

There are two types of signaling used in DVB-T2 system, one is L1 pre signaling and the other is L1 post signaling composed of configurable part, dynamic part and extension part. In FIG. 9, the method 2 is described based on L1 post signaling among T2 type signaling, but it is obvious that the method 2 may be applied to other L1 pre type signaling.

Hereinafter, a detailed operation of the present invention will be described based on DVB-NGH, which is a European next-generation mobile broadcasting system. However, the present invention is not limited to DVB-NGH.

As in Method 2, the signaling information is encoded, and non-punctured parity bits are transmitted in the same frame as the information word, and additional parity bits including the information word and parity bits are transmitted in the previous frame. At this time, the additional parity bits preferentially select the punctured parity bits, and the selection order is the order of the puncturing patterns or the reverse order.

개의 패리티 비트들(parity bits)을

개의 그룹으로 나눌 수 있으며 j번째 패리티 그룹

는 수학식 9와 같이 표현할 수 있다.given

Parity bits

J parity group

Can be expressed as in Equation (9).

는 상기 도 3에서 언급한 열 그룹을 구성하는 열의 개수

,

와

를 기반으로 하여 수학식 10과 같이 구할 수 있다. here

Is the number of columns constituting the column group mentioned in FIG.

,

Wow

Based on the equation (10) can be obtained.

개의 그룹으로 나누어지며 각 그룹은

개의 비트들로 구성된다. Therefore parity bits

Divided into groups, each group

It consists of three bits.

에 대하여 이하와 같이 천공되는 패리티 비트들을 구할 수 있다. The number of puncture bits given

For parity bits can be obtained as follows.

를 구할 수 있다. 이하 수학식 11에서의 연산

는 a를 넘지 않는 최대 정수를 의미한다. 예를 들어,

이다.Step 1: the number of groups in which all parity bits are punctured as shown in Equation 11

Can be obtained. Calculation in Equation 11 hereinafter

Is the maximum integer not exceeding a. E.g,

to be.

개의 패리티 비트 그룹들

은 그룹에 속한 모든 패리티 비트들이 천공된다. 상기

는 패리티 그룹의 인덱스를 나타내는 것으로 천공을 하는 순서를 의미한다. 예를 들어 표 1과 같이 정의할 수 있다. 단

가 표 1의 값에 국한된 것은 아니며, 본 명세서에서는 구체적인 값들은 언급하지 않도록 한다. 즉, 상기에서 언급한 천공 패턴을 의미하며

이다. 이는 상기 표1에서 설명한 바 있다. 상기

의 상세한 값은 변경이 가능하며, 이에 대한 상세한 설명은 본 발명의 논지를 흐릴 수 있으므로 생략한다.Step 2: above

Parity bit groups

Is punctured for all parity bits belonging to the group. remind

Denotes the index of the parity group, and indicates the order of puncturing. For example, it can be defined as shown in Table 1. only

Is not limited to the values in Table 1, and specific values are not mentioned herein. That is, it means the above-mentioned perforation pattern

to be. This has been described in Table 1 above. remind

The detailed value of may be changed, and a detailed description thereof will be omitted since it may obscure the subject matter of the present invention.

에 대하여서는 상기 그룹의 첫 번째 패리티 비트들부터

개의 비트들을 천공한다. 천공되지 않는 패리티 비트들은 정보어와 동일 프레임으로 전송한다.Step 3: Group

For the first parity bits of the group

Punctures two bits. Parity bits that are not punctured transmit in the same frame as the information word.

<Example 7>

Hereinafter, a method of selecting additional parity bits among the punctured bits will be described.

A method of obtaining the first group of additional parity bits G (0) is as follows. For convenience of explanation, it is assumed that the processes from step 4 are performed after the steps 1 to 3 are performed.

Step 4: The number of groups in which all elements in the parity bit group are selected to select additional parity bits is obtained as shown in Equation 12.

는 첫 번째 추가적인 패리티 비트들의 개수를 의미한다. From above

Denotes the number of first additional parity bits.

개의 패리티 비트 그룹들

은 그룹에 속한 모든 패리티 비트들은 첫 번째 추가적인 패리티 비트들 그룹 G(0)를 구성하며 정보어가 전송되는 프레임보다 앞선 프레임으로 전송된다. 상기 앞선 프레임이라 함은 도 14 또는 도 22를 통하여 설명한 바와 같이 다른 RF 채널에서도 적용할 수 있다.Step 5: above

Parity bit groups

All parity bits belonging to the group constitute the first additional parity bits group G (0) and are transmitted in a frame preceding the frame in which the information word is transmitted. The foregoing frame may be applied to other RF channels as described with reference to FIG. 14 or 22.

에 대하여서는 그룹 내의 첫 번째 패리티 비트들부터

개의 패리티 비트들은 첫 번째 추가적인 패리티 비트들 그룹 G(0)를 구성한다. Step 6: Group

For the first parity bits in the group

Parity bits constitute the first additional parity bits group G (0).

의 순서대로 패리티 비트들의 그룹

단위로 나열한다. 또는 도 12의 (b)에 도시한 바와 같이, G(0)로 선택된 패리티 비트들을 패리티 비트들의 인덱스 순서대로 비트 단위로 정렬할 수도 있다.There may be several ways to determine the order of the bits in the first group of additional parity bits G (0). For example, as shown in FIGS. 12A and 12C, a group of parity bits

Group of parity bits in the order of

List them in units. Alternatively, as shown in (b) of FIG. 12, parity bits selected by G (0) may be aligned bit by bit in the index order of the parity bits.

The following describes how to configure the second additional parity bits group G (1).

에서 첫 번째 추가적인 패리티 비트들 그룹을 구성하지 않은 패리티 비트들의 개수를 수학식 13과 같이 구할 수 있다.

와

에 대해서는 도 12에 도시하였다.Step 7: Group as below

In Equation 13, the number of parity bits that do not form the first additional parity bit group can be obtained.

Wow

It is shown in FIG.

에서

번째 비트부터 두 번째 추가적인 패리티 비트들 그룹 G(1)를 구성할 수 있다. 상기 a번째 비트라 함은 그룹

의 처음 비트를 0번째 비트라고 하였을 때 a번째에 위치한 비트를 의미한다. 또한 함수 min(a,b)는 a와 b중 작은 값을 선택하는 것을 의미하는 것으로, 만약 a≤b이면 min(a,b)=a이고 a>b이면 min(a,b)=b이다.Therefore groups

in

The second additional parity bits group G (1) may be configured from the first bit. The a-th bit is a group

When the first bit of is called 0th bit, it means the bit located in the ath. Also, the function min (a, b) means to select the smaller value of a and b. If a≤b, min (a, b) = a and if a> b, min (a, b) = b. .

인 경우, 추가적인 패티리 비트들을 선택하기 위하여 패리티 비트 그룹 내의 모든 원소들이 선택되는 그룹의 개수를 이하 수학식 14와 같이 구한다.Step 8: If

In the following case, the number of groups in which all elements in the parity bit group are selected in order to select additional parity bits is obtained as in Equation 14 below.

의 패리티 들은 모두 두 번째 추가적인 패리티 비트들의 그룹을 구성한다.Step 9: Group

The parities of make up the second group of additional parity bits.

에서는

개의 패리티 비트들이 두 번째 추가적인 패리티 비트들로 구성된다.Step 10: Group

In

Parity bits are composed of second additional parity bits.

 The following describes how to configure the third additional parity bits group G (2).

에서 두 번째 추가적인 패리티 비트들 그룹을 구성하지 않은 패리티 비트들의 개수

를 수학식 15와 같이 구할 수 있다.Step 11: Group

Number of parity bits that do not form a second group of additional parity bits in

May be obtained as shown in Equation 15.

에서

번째 비트부터 y개의 비트들을 세 번째 추가적인 패리티 비트들 그룹 G(2)를 구성할 수 있다.Therefore groups

in

From the first bit to y bits may constitute a third additional parity bit group G (2).

인 경우, 세 번째 추가적인 패리티 비트들을 선택하기 위하여 패리티 비트 그룹 내의 모든 원소들이 선택되는 그룹의 개수를 수학식 16과 같이 구한다.Step 12: If

In Equation 16, the number of groups in which all elements in the parity bit group are selected in order to select third additional parity bits is obtained as in Equation 16 below.

의 패리티들은 모두 세 번째 추가적인 패리티 비트들의 그룹을 구성한다.Step 13: Group

The parities of all make up the third group of additional parity bits.

에서는

개의 패리티 비트들이 두 번째 추가적인 패리티 비트들로 구성된다.Step 14: Group

In

Parity bits are composed of second additional parity bits.

The method of determining the order of bits in the group G (1) of the second additional parity bits is the same as the method of determining the order of bits in the group G (1) of the first additional parity bits.

The order in which the bits of groups G (0) through G (1) of additional parity bits are transmitted on air may be obscured by the subject matter of the present invention.

<Example 8>

Hereinafter, a method of selecting the reverse parity pattern in the method of selecting additional parity bits will be described with reference to FIG. 13.

The method for obtaining the parity bits transmitted in the same frame as the information word is the same as the above steps 1 to 3.

A method of obtaining the first additional parity bits group G (0) is as follows.

에서 그룹 G(0)으로 선택되는 비트들의 개수

를 구할 수 있다. Step 4: Group as shown in Equation 17

Number of bits selected as group G (0) in

Can be obtained.

에서는 그룹 내의 첫 번째 비트부터

개의 비트들을 첫 번째 추가적인 패리티 비트들로 선택한다.Step 5: Group

From the first bit in the group

Bits are selected as the first additional parity bits.

에 대하여서는 상기 그룹의 마지막 패리티 비트들부터

개의 비트들을 천공하고, 천공되지 않는 비트들은 정보어와 동일 프레임으로 전송한다면 상기 단계　5에서 그룹

에서는 그룹 내의 마지막 패리티 비트부터

개의 비트들을 첫 번째 추가적인 패리티 비트들로 선택한다.If group in step 3 above

For, from the last parity bits of the group

If the four bits are punctured and the non-punctured bits are transmitted in the same frame as the information word,

Is the last parity bit in the group.

Bits are selected as the first additional parity bits.

가

보다 클 경우 이하의 과정을 수행한다. if

end

If larger, perform the following procedure.

Step 6: Obtain the number of groups of parity bits in which all bits in the group are transmitted as shown in Equation 18.

개의 패리티 비트 그룹들

,

,...,

은 그룹에 속한 모든 패리티 비트들은 첫 번째 추가적인 패리티 비트들 그룹 G(0)를 구성하며 정보어가 전송되는 프레임보다 앞선 프레임으로 전송된다. Step 7: Remind

Parity bit groups

,

, ...,

All parity bits belonging to the group constitute the first additional parity bits group G (0) and are transmitted in a frame preceding the frame in which the information word is transmitted.

에 대해서는 그룹 내의 첫 번째 패리티 비트들부터

개의 패리티 비트들은 첫 번째 추가적인 패리티 비트들 그룹 G(0)를 구성한다. Step 8: Group

For, from the first parity bits in the group

Parity bits constitute the first additional parity bits group G (0).

의 역순으로 패리티 비트 그룹

단위로 나열한다. 또는 도 13의 (b)에 도시한 바와 같이, 추가적인 패리티 비트들로 선택된 패리티 비트들을 패리티 비트들의 인덱스 순서대로 비트 단위로 정렬할 수도 있다. There may be several ways to determine the order of the bits in the first group of additional parity bits G (0). For example, as shown in FIGS. 13A and 13C, a group of parity bits

Group of parity bits in reverse order

List them in units. Alternatively, as shown in (b) of FIG. 13, parity bits selected as additional parity bits may be arranged in bit units in the index order of parity bits.

Hereinafter, a second additional parity bit group G (1) is configured.

에서 그룹 G(1)으로 선택되는 비트들의 개수

를 구할 수 있다. Step 9: Group as shown in Equation 19

Number of bits selected as group G (1) in

Can be obtained.

에서는 그룹 내의 처음 비트부터

개의 비트들을 선택한다. Step 10: Group

From the first bit in the group

Bits are selected.

가

보다 클 경우 이하의 과정을 수행한다.if

end

If larger, perform the following procedure.

Step 11: Find the number of groups in which all the bits in the group are transmitted as shown in Equation 20.

개의 패리티 비트 그룹들

,

,...,

은 그룹에 속한 모든 패리티 비트들은 두 번째 추가적인 패리티 비트들 그룹 G(1)를 구성하며 정보어가 전송되는 프레임보다 앞선 프레임으로 전송된다. Step 12: above

Parity bit groups

,

, ...,

All parity bits belonging to the group constitute a second additional parity bit group G (1) and are transmitted in a frame preceding the frame in which the information word is transmitted.

에 대하여서는 그룹 내의 첫 번째 패리티 비트들부터

개의 패리티 비트들은 두 번째 추가적인 패리티 비트들 그룹 G(1)를 구성한다. Step 13: Group

For the first parity bits in the group

Parity bits constitute a second additional parity bit group G (1).

The method of determining the order of bits in the group G (1) of the second additional parity bits is the same as the method of determining the order of bits in the group G (1) of the first additional parity bits.

The order in which the bits in groups G (0) to G (1) of additional parity bits are transmitted on air is obscure, as this may obscure the subject matter of the present invention.

Example 9

Hereinafter, a method of more efficiently obtaining additional parity bits by interleaving the parity bits will be described with reference to FIG. 14.

에 대하여 수학식 21과 같이 인터리빙을 한다.Parity Bits Obtained by LDPC Coding

The interleaving is performed as in Equation 21.

를 기반으로 하여 그룹 인터리빙을 하며 수학식 22와 같이 나타낼 수 있다.As mentioned above with respect to the parity bit group

The group interleaving is performed based on, and can be expressed as Equation 22.

단위로 정렬된다.When interleaving as shown in Equation 21, the LDPC codeword configured as shown in FIG. 14A has a plurality of parity bit groups as shown in FIG. 14B.

Sorted by unit.

의 비트들은

의 비트들과 동일하다. 상기

는 천공을 하는 순서를 나타내는 값으로 상기에서 언급한 천공 패턴을 의미 하며

이다. 이는 상기 표1에서 설명한 바 있다. 상기

의 상세한 값은 변경이 가능하여 상세한 설명은 본 발명의 논지를 흐릴 수 있으므로 생략한다.The group interleaving means interleaving in group units, which means that bits in the group are equally interleaved.

Bits of

Is the same as the bits of. remind

Is a value indicating the order of drilling and means the above-mentioned drilling pattern.

to be. This has been described in Table 1 above. remind

The detailed value of can be changed so that the detailed description will be omitted since it may obscure the subject matter of the present invention.

When interleaving as shown in Equation 22, the LDPC codeword configured as shown in FIG. 14 (b) is arranged in the reverse order of the puncturing pattern as shown in FIG. 14 (c).

Hereinafter, a method of obtaining parity bits transmitted in the same frame as the information word will be described.

를 구할 수 있다.Step 1: the number of groups in which all parity bits are transmitted based on the number of parity bits to be punctured as shown in Equation 23

Can be obtained.

개의 패리티 비트 그룹들

은 그룹에 속한 모든 패리티 비트들이 전송된다. Step 2: above

Parity bit groups

All parity bits belonging to the group are transmitted.

에 대하여서는 그룹의 첫 번째 패리티 비트들부터

개의 비트들을 정보어와 동일 프레임으로 전송한다.Step 3: Group

For the first parity bits of the group

Bits are transmitted in the same frame as the information word.

The method of obtaining the first additional parity bits is as follows.

에서 그룹 G(0)로 선택되는 비트들의 개수

를 구할 수 있다. Step 4: Group as shown in Equation 24

Number of bits selected as group G (0) in

Can be obtained.

에서는 그룹 내의 첫 번째 비트부터

개의 비트들을 첫 번째 추가적인 패리티 비트들로 선택한다. Step 5: Group

From the first bit in the group

Bits are selected as the first additional parity bits.

가

보다 클 경우 이하의 과정을 수행한다.if

end

If larger, perform the following procedure.

Step 6: Obtain the number of groups in which all the bits in the group are transmitted as shown in Equation 25.

개의 패리티 비트 그룹들

,...

은 그룹에 속한 모든 패리티 비트들은 첫 번째 추가적인 패리티 비트들 그룹 G(0)를 구성하며 정보어가 전송되는 프레임보다 앞선 프레임으로 전송된다. Step 7: Remind

Parity bit groups

, ...

All parity bits belonging to the group constitute the first additional parity bits group G (0) and are transmitted in a frame preceding the frame in which the information word is transmitted.

에 대하여서는 그룹 내의 첫 번째 패리티 비트들부터

개의 패리티 비트들은 첫 번째 추가적인 패리티 비트들 그룹 G(0)를 구성한다. Step 8: Group

For the first parity bits in the group

Parity bits constitute the first additional parity bits group G (0).

As mentioned above, a group of additional parity bits is configured as shown in FIG. 14 (d).

Hereinafter, a second additional parity bit group G (1) is configured.

에서 그룹 G(1)로 선택되는 비트들의 개수

를 구할 수 있다. Step 9: Group as shown in Equation 26

Number of bits selected by group G (1) in

Can be obtained.

에서는 그룹 내의 처음 비트부터

개의 비트들을 선택한다. Step 10: Group

From the first bit in the group

Bits are selected.

가

보다 클 경우 이하의 과정을 수행한다.if

end

If larger, perform the following procedure.

Step 11: Obtain the number of groups in which all the bits in the group are transmitted as shown in Equation 27.

개의 패리티 비트 그룹들

,

,...,

은 그룹에 속한 모든 패리티 비트들은 두 번째 추가적인 패리티 비트들 그룹 G(1)를 구성하며 정보어가 전송되는 프레임보다 앞선 프레임으로 전송된다. Step 12: above

Parity bit groups

,

, ...,

All parity bits belonging to the group constitute a second additional parity bit group G (1) and are transmitted in a frame preceding the frame in which the information word is transmitted.

에 대하여서는 그룹 내의 첫 번째 패리티 비트들부터

개의 패리티 비트들은 두 번째 추가적인 패리티 비트들 그룹 G(1)를 구성한다.
Step 13: Group

For the first parity bits in the group

Parity bits constitute a second additional parity bit group G (1).

Group interleaving of Equation 22 may be expressed as Equation 28.

로 나타낼 수 있다. In this case the parity bits

It can be represented as.

값과 추가적인 패리티 비트들인

을 기반으로 추가적인 패리티 비트들을 사용하지 않을 경우에도 선택되는 패리티 비트들을 선택할 수 있다.Therefore, the information transmitted in the same frame as the information word obtained in Equation 23

Value and additional parity bits

Based on this, even when no additional parity bits are used, parity bits selected may be selected.

이 외에도 본 발명을 구현하는 방법에는 여러 가지가 존재할 수 있음은 당연하다.

In addition, there are many ways to implement the present invention may be present.

이 천공되는 패리티 비트들의 개수

보다 작은 경우에 대하여 설명하였다. 그런데 만약 상기 추가적인 패리티 비트들의 개수의 총합

이

보다 클 경우

비트 만큼에 대해서는 상기 실시예를 따르며, 추가적인 패리티 비트들이 더 필요한 경우에는 상기 정보어 및 패리티 비트들 및 추가적인 패리티 비트들을 순차적으로 반복하여 선택한다. 이를 도 23에 간단히 도시 하였다.
In the above, the sum of the number of additional parity bits

The number of these perforated parity bits

A smaller case has been described. But if the sum of the number of additional parity bits

this

Greater than

As for the bits, the above-described embodiment is used. When additional parity bits are required, the information word and the parity bits and the additional parity bits are sequentially selected repeatedly. This is shown briefly in FIG.

<Example 10>

이

보다 클 경우에 대하여 고려하도록 한다.Hereinafter, a method of selecting the additional parity bits in the reverse order of the puncturing pattern will be described. Specifically the sum of the number of additional parity bits

this

Consider larger cases.

The method for obtaining the parity bits transmitted in the same frame as the information word is the same as the above steps 1 to 3.

A method of obtaining the first additional parity bits group G (0) is as follows.

인 경우

비트들을 그룹 G(0)로 선택하고,

비트에 대해서는 부호어 중에서 G(0)를 선택한다. 이때 가장 간단한 방법은 부호어 및 추가적인 패리티 비트들을 순차적으로 반복하여 선택하는 것이다. 이를 도 23에 간단히 도시 하였다.Step 4: If

If

Select bits into group G (0),

For the bit, G (0) is selected from the code words. In this case, the simplest method is to sequentially select a codeword and additional parity bits. This is shown briefly in FIG.

인 경우, 상기 실시예 8의 단계４ 내지 단계 8을 동일하게 수행한다.if

In the case of, step 4 to step 8 of the eighth embodiment is performed in the same manner.

A method of configuring the second additional parity bits group G (1) is as follows.

인 경우 부호어 중에서 G(1)를 선택한다. 이때 가장 간단한 방법은 이전에 전송했던 비트들을 순차적으로 선택하는 것이다. Step 9: If in step 4 above

In case of, G (1) is selected from codewords. In this case, the simplest method is to sequentially select previously transmitted bits.

이고

인 경우, 부호어 중에서 G(1)를 선택한다. 이때 가장 간단한 방법은 상기 정보어 및 패리티 비트들 및 추가적인 패리티 비트들을 순차적으로 반복하여 선택하는 것이다. 이를 도 23에 간단히 도시하였다.if

ego

In the case of, G (1) is selected from codewords. In this case, the simplest method is to sequentially select the information word, the parity bits, and the additional parity bits. This is shown briefly in FIG.

이면, 상기 실시예 8의 단계 9 내지 단계 13과 동일하다.if

In this case, the process is the same as the steps 9 to 13 of the eighth embodiment.

내의 비트들이 전송되는 순서에 대해서는 본 발명의 논지를 흐릴 수 있으므로 생략한다. Group above

The order in which the bits within are transmitted are omitted because they may obscure the subject matter of the present invention.

In addition, there are many ways to implement the present invention may be present.

15 is a block diagram of a communication system capable of obtaining the diversity effect proposed in the present invention.

Referring to FIG. 15, the message u is encoded through an encoder 1502 of the transmitter 1500, modulated by a modulator 1504, and framed by the frame configurator 1506 before being transmitted. This is configured and then transmitted over channel 1508. The transmitted signal is received by the receiver 1510 and deconfigured by the frame deconfigurer 1516 and input to the demodulator 1514, and the decoder 1533 is demodulated by the demodulator 1514. Estimate the estimate of the message from the signal.

The encoder 1502 generates parity bits using puncturing and shortening according to the size of the message from a preset method.

A specific example is shown in FIG. 16 to show a transmission apparatus for realizing a shortening process of the LDPC code in more detail. 16 is a block diagram of a transmitter using a shortened LDPC code according to an embodiment of the present invention.

The transmitting device includes an encoder 1602, a controller 1604, a perforator 1606, an additional parity bit group generator 1608, and a frame configurator 1610, and may optionally include a shortening application unit 1600. have.

The controller 1604 controls the shortening applying unit 1600 to determine the number of bits to be shortened according to the length of the information word, and the shortening applying unit 1600 has a bit having a zero value at a position corresponding to the shortened bit. Insert or remove the column corresponding to the shortened bit from the parity check matrix of the given LDPC code. As a method of determining the shortened pattern, a shortened pattern stored using a memory is used, a shortened pattern is generated using a sequence generator (not shown), or the density of the parity check matrix and a given information word length is determined. And methods obtained using an evolutionary analysis algorithm.

The LDPC encoder 1602 performs encoding based on the LDPC code shortened by the control unit 1604 and the shortening application unit 1600. In addition, when proper puncturing is required, puncturing is applied to the LDPC codeword generated by the puncturing unit 1606. The number of bits to be punched and the position of the punch in the puncher 1606 are determined by the controller 1604 according to the punch pattern. That is, the controller knows the puncturing pattern which is the order of puncturing and selects the punctured bits according to the order. The perforation pattern may be stored in a memory or may generate perforation patterns using a sequence generator (not shown).

The additional parity bit group generator 1608 receives the output data of the controller 1604 and the encoder 1602 and the data of the puncturer 1606 and generates additional parity bit groups according to the above-described rules.

The frame configurator 1610 configures a frame corresponding to the methods 1 to 2 by the puncturer 1606, the additional parity bit group generator 1608, and the controller 1604.

17 is a block diagram of a receiving apparatus using an LDPC code to which an additional parity bit group is applied according to an embodiment of the present invention.

In FIG. 17, when a user receives a signal transmitted from a communication system using the additional parity bit group, and knows information of additional parity bits constituting the additional parity bit group from the received signal, the user may receive information from the received signal. An example of a receiving device for restoring desired data is shown.

The receiving apparatus includes a control unit 1700, a shortening or puncturing processor 1702, a demodulator 1704, an additional parity bit group processor 1706, and a decoder 1708.

The demodulator 1704 receives and demodulates an LDPC code, reconstructs a frame by a controller, and estimates an input value of the LDPC coder of FIG. 16 based on the received signal. The demodulated signal is passed to the shortening or puncturing processor 1702 and the decoder 1708 and the additional parity bit group processor 1706.

The shortening or puncturing processor 1702 determines information on shortened or punctured bits of the LDPC code from the signal demodulated by the demodulator 1704 under the control of the control unit 1700, and the shortened and punctured bits. The location information of the is transmitted to the decoder 1708.

The decoder 1708 obtains data desired by the user from the received signal using the output value of the demodulator 1704 and the length of the shortened and punctured codes input from the shortened or punctured processor 1702 and the position information of the corresponding bits. Can be restored The decoder 1708 may also receive position information and a demodulation value of the additional parity bits from the additional parity bit group processor 1706 and decode them.

The additional parity bit group processor 1706 processes and transmits the location of the additional parity bits and the demodulated data to the decoder 1708 under the control of the controller 1700. The meaning of processing the demodulated data may vary. For example, when the same bit is received several times, the demodulated value is processed.

Another specific example is shown in FIG. 18 to more specifically show a transmitting apparatus for realizing a shortening process of the LDPC code. 18 is a block diagram of a transmitter using a shortened LDPC code according to an embodiment of the present invention.

The transmitting apparatus includes an encoder 1802, a controller 1804, a perforator 1806, an additional parity bit group generator 1808, a frame configurator 1810, and a parity interleaver 1812, and in some cases, a shortening application unit ( 1800).

The controller 1804 controls the shortening applying unit 1800 to determine the number of bits to be shortened according to the length of the information word, and the shortening applying unit 1800 has a bit having a zero value at a position corresponding to the shortened bit. Insert or remove the column corresponding to the shortened bit from the parity check matrix of the given LDPC code. As a method of determining the shortened pattern, a shortened pattern stored using a memory is used, a shortened pattern is generated using a sequence generator (not shown), or the density of the parity check matrix and a given information word length is determined. And methods obtained using an evolutionary analysis algorithm.

The encoder 1802 performs encoding based on the LDPC code shortened by the control unit 1804 and the shortening application unit 1800. In addition, the output values of the encoder 1802 are input to the parity interleaver 1812 to perform parity interleaving as mentioned in Embodiment 9. In this case, parity interleaving may be performed in units of groups, or interleaving may be performed in units of bits. As mentioned above, the parity interleaver 1812 is interleaved by the controller 1808 according to the puncturing pattern. In addition, when proper puncturing is required, puncturing is applied to the LDPC codeword generated by the puncturing unit 1806. The number of bits to be punched and the position of the punch in the puncher 1806 are determined by the controller 1804 according to the punch pattern. That is, the controller knows the puncturing pattern which is the order of puncturing and selects the punctured bits according to the order. The perforation pattern may be stored in a memory or may generate perforation patterns using a sequence generator (not shown).

The additional parity bit group generator 1808 receives the output data of the controller 1804 and the encoder 1802 and the data of the puncturer 1806 and generates additional parity bit groups according to the above-described rules.

The frame configurator 1810 configures a frame corresponding to the methods 1 to 2 by the puncturer 1806, the additional parity bit group generator 1808, and the controller 1804.

19 is a block diagram of a receiving apparatus using an LDPC code to which an additional parity bit group is applied according to an embodiment of the present invention.

In FIG. 19, when a user receives a signal transmitted from a communication system using the additional parity bit group and learns information of additional parity bits constituting the additional parity bit group from the received signal, the user may receive information from the received signal. An example of a receiving device for restoring desired data is shown.

The receiving apparatus includes a controller 1900, a shortening or puncturing processor 1902, a demodulator 1904, an additional parity bit group processor 1906, a decoder 1908, and a parity deinterleaver 1910.

The demodulator 1904 receives and demodulates an LDPC code, and transfers the demodulated signal to a shortened or puncture processor 1902, a decoder 1908, and an additional parity bit group processor 1906.

The abbreviation or puncturing processor 1902 determines the information on the shortened or punctured bits of the LDPC code from the signal demodulated by the demodulator 1904 under the control of the controller 1900, and the shortened bits and the punctured bits. The position information of the bit is transmitted to the decoder 1908.

The parity deinterleaver 1910 is a corresponding block of the parity interleaver 1812 of FIG. 18, which outputs the demodulator 1904 and the length of the shortened and punctured code input from the shortened or punctured processor 1902 and the corresponding block. Parity bits are reconstructed using the position information of the bits and parity deinterleaving is performed.

The decoder 1908 uses the output value of the parity deinterleaver 1910, the length of the shortened and punctured code input from the shortened or punctured processor 1902, and the position information of the corresponding bits to the user. You can restore the data. The decoder 1908 may also receive position information and a demodulation value of additional parity bits from the additional parity bit group processor 1906 and decode them.

The additional parity bit group processor 1906 processes and transmits the location of the additional parity bits and the demodulated data to the decoder 1908 under the control of the controller 1900. The meaning of processing the demodulated data may vary. For example, when the same bit is received several times, the demodulated value is processed.

20 is a flowchart of a transmitter according to an embodiment using the frame structure and transmission method of the present invention.

The controllers 1604 and 1804 determine the number of bits for puncturing and shortening, the number of frames for transmitting parity bits in a plurality of frames, and the number and group of bits transmitted in each frame according to the method and rule described in operation 2000. Decide

The abbreviation application unit 1600 or 1800 may shorten the process if necessary in step 2002. The encoders 1602 and 1802 perform LDPC encoding with the parameters determined in step 2004. The punchers 1606 and 1806 shorten and puncture the bits encoded in step 2006. In step 2008, the additional parity bit group generators 1608 and 1808 generate a plurality of additional parity bit groups according to the rules 3 to 5 for the codeword bits and the parity bits. Alternatively, this process may include a parity interleaving process. The frame organizers 1610 and 1810 transmit codewords and a plurality of additional parity groups in various methods in a plurality of frames according to the above-described embodiments in step 2010.

21 is a flowchart illustrating a receiving operation in a receiving apparatus according to an embodiment of the present invention.

The receiver receives a sign in step 2100. Demodulators 1704 and 1904 demodulate the signals received in step 2102. The shortened or punctured processors 1702 and 1902 then process the shortened or punctured values from the demodulated signal in step 2104.

If there are no shortened or punctured bits, the decoders 1708 and 1908 perform decoding in step 2108 without performing step 2106. However, if there is a shortened or punctured bit, the shortened or punctured processors 1702 and 1902 transfer the position information of the shortened / punched bits to the decoders 1708 and 1908 in step 2106, and the additional parity bit group processor 1706. 1906 passes the location information of the additional parity bits to the decoders 1708 and 1908.

The decoders 1708 and 1908 determine that the value of the shortened bit is 1, based on the location information of the shortened / punched bit in step 2108, and the punctured bit is an erased bit. After the determination, the LDPC decoding is performed.

Claims (25)

In a method for transmitting data in a communication system,
Transmitting the information word constituting the codeword in the k + s-th frame,
Generating s groups based on parity bits obtained by encoding the information words;
And dividing the s groups into s frames, each of which precedes a k + s-th frame. The method of claim 1,
The process of creating the group,
And generating the remaining bits except the bits punctured according to a predetermined puncturing pattern among the parity bits into s groups in order according to the puncturing pattern or in reverse order of the puncturing pattern. The method of claim 2,
And mapping bits contiguous in the puncturing pattern among the remaining bits except the punctured bits to the same group. The method of claim 1,
The process of creating the group,
And mapping the remaining bits except the bits punctured according to a predetermined puncturing pattern among the parity bits into s groups, respectively. The method of claim 4, wherein
And mapping contiguous bits among the remaining bits except the punctured bits to different groups. The method of claim 5,
And the parity bits are a double diagonal structure. The method according to claim 2 or 4,
The process of creating the group,
And after selecting all the remaining bits, generating the group by sequentially selecting the punctured bits according to the puncturing pattern. The method of claim 7, wherein
The process of creating the group,
And generating a group by sequentially selecting bits of the codeword after selecting all of the parity bits when the total number of bits forming the group is larger than the number of the parity bits. In a method for transmitting data in a communication system,
Transmitting the information word constituting the code word and the remaining bits except for the bits punctured according to a predetermined puncturing pattern among the parity bits obtained by encoding the information word in a k + s th frame;
Generating the s groups by selecting the punctured bits sequentially or in the reverse order of the puncture pattern according to the puncture pattern;
And dividing the s groups into s frames, each of which precedes a k + s-th frame. 10. The method of claim 9,
The process of creating the group,
And after selecting all the punctured bits, sequentially selecting the bits of the information word to generate the s groups. The method of claim 10,
The process of creating the group,
And after selecting all of the information word bits, the remaining bits are sequentially selected according to the puncturing pattern or in the reverse order of the puncturing pattern to generate the s groups. 10. The method of claim 9,
And transmitting parity bits adjacent to each other in the puncturing pattern to the same group. An apparatus for transmitting data in a communication system,
An encoder for encoding the information word according to a predetermined method,
A puncturer for puncturing the codeword coded by the encoder according to a predetermined puncturing pattern;
A parity group generator for generating parity bits punctured by the puncturer among the codewords output from the encoder into s groups;
And a transmitter for dividing and transmitting the information word constituting the codeword and the group of parity bits into k + s-th frames and s frames preceding the k + s-th frames, respectively. The method of claim 13,
The parity group generator,
And mapping the remaining bits excluding the bits punctured according to a predetermined puncturing pattern among the parity bits to the s groups in order according to the puncturing pattern or in reverse order of the puncturing pattern. The method of claim 14,
The parity group generator,
And a bit concatenated in the puncturing pattern among the remaining bits except the punctured bits to the same group. The method of claim 13,
The parity group generator,
And a plurality of bits except for the bits punctured according to a predetermined puncturing pattern among the parity bits in order to s groups. The method of claim 16,
The parity group generator,
And a concatenated bit among the remaining bits except for the punctured bits to different groups. The method of claim 17,
And the parity bits have a dual diagonal structure. The method according to claim 14 or 16,
The parity group generator,
And selecting all the bits that are punctured sequentially according to the puncturing pattern after generating all the remaining bits to generate the group. The method of claim 19,
The parity group generator,
And generating a group by sequentially selecting the bits of the codeword after selecting all of the parity bits when the total number of bits forming the group is larger than the number of parity bits. The method of claim 13,
The transmitter is a data transmission apparatus for transmitting the information word and the remaining bits, except the bits punctured according to a predetermined puncturing pattern among the parity bits, preferentially in a k + s-th frame. The method of claim 21,
The parity group generator,
And generating the s groups by selecting the punctured bits in order according to the puncturing pattern or in reverse order of the puncturing pattern. The method of claim 22,
The parity group generator,
And after selecting all of the punctured bits, sequentially selecting the bits of the information word to generate the s groups. The method of claim 22,
The parity group generator,
And after selecting all the information word bits, the remaining bits are sequentially selected according to the puncturing pattern or in the reverse order of the puncturing pattern to generate the s groups. The method of claim 22,
The parity group generator,
And transmitting parity bits adjacent to each other in the puncturing pattern to the same group.

Images