KR20080104684A - Method for puncturing of block type low density parity check code - Google Patents

Abstract

A method for punching a block type low density parity check in a bit unit is provided to obtain high decoding performance by punching bit uniformly when connecting a zigzag edge. A communication system supporting a variable code rate includes a code rate controller(S121), a punching position operator(S122,S172), a transmitter(S140), and a receiver(S150). The code rate controller delivers the information required for punching. The punching position operator calculates the position of the punching bit by using the information connected to the code rate controller. The punching position operator receives the information in the code rate controller through the control channel and calculates the position of the punching bit. A transmitter is connected to the punching position operator. The transmitter includes a coding bit selector(S123) which selects the bit which is not punched among the coded bit inputted to the punching device and delivers the bit. The receiver is connected to the punching position operator. A coding bit distribution and 0 inserter(S171) distributes temporary determining variable of the coding bit unit and inserts 0 into the position of the punching bit, and outputs the punching bit.

Description

Method for Puncturing Of Block Type Low Density Parity Check Code}

1 is a view showing a bit position search order for puncturing according to the present invention;

2 is a diagram illustrating a configuration of a transceiver for supporting variable code rate according to the present invention;

3 is a diagram illustrating a configuration of a transceiver for HARQ operation according to the present invention;

4 is a view showing a procedure for selecting a block to be drilled according to the present invention;

5 is a flow chart showing a drilling method according to the present invention,

6 is a flow chart that shows the punching method according to the present invention formulaally,

7 is a diagram schematically illustrating selection of a block unit and a bit unit to be punctured using the flowchart of FIG. 6;

8 is a graph showing the effectiveness of bit unit selection in the drilling method;

9 is a graph showing the performance of a punctured code by applying a puncture algorithm in a block LDPC mother code having a code rate of 1/2;

10 is a graph showing the performance of a code punctured by applying a puncturing algorithm in a block LDPC mother code having a code rate of 1/3.

The present invention relates to a method for puncturing a block type low density parity check code, and more particularly, in a block type low density parity check code having a bi-diagonal parity block structure to maximize the speed of the minimum recovery speed among the puncturing nodes. By drilling the parity bit nodes and arranging the puncture nodes as uniformly as possible in the zigzag edge connection, it minimizes performance degradation due to puncturing.Effective puncture in bits even when the number of nodes to be punctured is not an integer multiple of the block size The present invention relates to a drilling method that provides a pattern and effectively uses Hybrid Automatic Repeat and reQuest (HARQ) technology.

In general, as the mobile communication system develops rapidly, there is a demand for developing a technology capable of transmitting a large amount of data approaching the capacity of a wired network in a wireless network. In this way, as a high-speed mass communication system capable of processing and transmitting various information such as video and wireless data beyond voice-oriented services is required, improving system transmission efficiency by using an appropriate channel coding method is improved system performance. It becomes an essential element for. However, due to the characteristics of the mobile communication system, the mobile communication system inevitably generates an error due to noise, interference, fading, etc. according to channel conditions when data is transmitted. The loss of the information data due to the error occurs.

Therefore, there is a need for an encoder and decoder having high performance in order to recover information transmitted from a transmitter without error in a receiver. In particular, due to the characteristics of the mobile communication system, the radio channel environment must be taken into consideration, so that an error caused by the radio channel environment must be considered more seriously.

In order to reduce information data loss due to such an error, reliability of the mobile communication system may be improved by using various error control schemes according to the characteristics of the channel. Among the error control schemes, the most commonly used error control scheme is an error-correcting code.

Representative codes of the error correction code include a turbo code, a low density parity check (LDPC) code, and the like. The turbo code is known to have a superior performance gain in high-speed data transmission, compared to a convolutional code used mainly for error correction. The turbo code effectively corrects errors due to noise generated in a transmission channel and transmits data. It has the advantage of increasing the reliability of. In addition, the LDPC code can be decoded using an iterative decoding algorithm based on a sum-product algorithm on a factor graph.

LDPC was first proposed by Gallager in 1962, and is a linear block code in which most of the elements of the parity check matrix H are '0'. Has long been forgotten. However, Mackay and Neal rediscovered it and showed excellent performance using Gallager's simple probabilistic decoding method.

The LDPC code is defined by a random parity check matrix H having a small number of '1's in the matrix. The parity check matrix H is a matrix for checking whether the received signal is normally decoded. When the product of the encoded received signal and the parity check matrix H becomes '0', it is determined that no error occurs. Accordingly, the LDPC code first designs a parity check matrix, which may be '0' when multiplied for all the encoded received signals, and then performs an operation of encoding by the encoder on the transmitting side according to the determined parity check matrix. Inversely, it is calculated.

The LDPC code has advantages such as superior performance, low decoding complexity, and parallel processing, compared to a turbo code, but has a problem of requiring a relatively high coding complexity and a large amount of memory for storing a generation matrix in an encoder.

In order to solve this problem, a block-type LDPC (B-LDPC) code that has a simple complexity that is simply proportional to the length of a codeword and can be efficiently encoded has been proposed. This B-LDPC code can be represented with very small amount of memory and has the advantage that it can be easily changed with various codeword lengths, and it is comparable to well-designed LDPC code rather than block structure in performance. Because of this advantage, it has been adopted as an optional technology in standardization technologies such as IEEE 802.16e (WiMax).

However, these codes still have a disadvantage in that they cannot support various code rates compared to turbo codes. In order to support various code rates, it must have multiple parity check matrices. In addition, the Hybrid ARQ technology of the surplus incremental (IR) method is not currently supported in 802.16e.

In order to solve the above problems, an effective puncturing technique of the B-LDPC code is required. Recently, some techniques for drilling LDPC codes designed in any manner have been proposed. Although they can obtain a good puncturing pattern, it is difficult to obtain a puncturing pattern in real time due to a complicated puncturing algorithm, and a lot of memory is required to store an arbitrary puncturing pattern. Recently, a puncturing pattern for several B-LDPC codes has been proposed, but since it was not a specific and general puncturing pattern, but only a block puncturing structure, it is meaningless when a bit puncturing is required. do. Since then, a bitwise puncturing method has been proposed, but it was a simple method that did not properly utilize the bit substitution structure for each block of the actual B-LDPC code.

The present invention has been conceived in view of the above, it is possible to obtain a puncturing pattern in real time using less memory, and provides an efficient puncturing pattern in bits even when the number of puncture nodes is not an integer multiple of the block size. The purpose of this invention is to provide a drilling method that can effectively use the HARQ (Hybrid Automatic Repeat and reQuest) technology.

In the above-described puncturing method of the block-type low density parity check code of the present invention,

In the block-type low density parity check code having a bi-diagonal parity block structure,

Determining a predetermined block group;

Positioning a block within the group;

And determining a position of a bit to be punctured in the block.

In the determining of the predetermined block group, the punctured node may be recovered by repeated decoding of the smallest number of times.

In addition, the step of determining the predetermined block group, characterized in that the grouping so that the maximum number of k-SR node having the smallest k value.

In addition, the step of positioning the block in the group,

Checking whether the number of remaining blocks is odd;

And determining whether there are any blocks left unperforated in the current group.

In addition, the step of checking whether the number of remaining blocks is odd is made by checking whether m _b / 2 ^t is an integer, wherein the value of m _b is the number of rows of blocks, and t represents a group index. .

In addition, the step of positioning the bits to be punctured in the block,

The number of the remaining blocks the even and to determine the drilling bit positions by, left L ₂ (i) if the number is an odd number of blocks by L ₁ (i) If the block remains without being punctured in the current group, L ₁ (i) and L ₂ (i) are each

이며, 상기

는

그리고

는 c 보다 작거나 같은 최대 정수를 의미하는 것을 특징으로 한다.

And said

Is

And

Is a maximum integer less than or equal to c.

In addition, L ₁ (i) and L ₂ (i) are each

And u ' _z (i) and u'' _z (i) are each

And u _z (i) is

It is characterized by that.

The terminology used herein is for the purpose of describing particular example embodiments only and is not intended to be limiting of the present invention. A singular expression includes a plural expression unless the context clearly indicates otherwise. In this application, the terms “comprises” or “having” are intended to indicate that there is a feature, number, step, action, component, part, or combination thereof described in the specification, and one or more other It is to be understood that the present invention does not exclude the possibility of the presence or the addition of features, numbers, steps, operations, components, parts, or a combination thereof.

Hereinafter, the present invention will be described in detail with reference to the accompanying drawings.

The present invention applies to all B-LDPC codes having a block pair diagonal parity structure. The puncturing method of the present invention is based on a B-LDPC code having a low density parity check matrix including an information region and a parity region composed of a predefined number of blocks, and a predetermined block group ( According to S10, the position S20 of the block is determined, and the position S30 of the bit to be punctured within the determined block is determined.

Hereinafter, a structure of a transmitter / receiver using a puncturer and a divider according to the present invention to support variable code rates will be described with reference to FIG. 1.

First, the information bits are received, and the encoded bits encoded and output by the encoder S110 are input to the puncturer S120. The encoding bit selector S123 of the puncturer selects only the encoding bits to be transmitted by the puncturing position calculator S122. At this time, the puncturing position calculator receives the information about the length of the parent code and the puncturing amount according to the code rate to be transmitted from the code rate controller (S121) and performs the puncturing position calculation in real time to inform the selector of the bits to be punctured. The selector S123 selects only bits that are not punctured among all coded bits. The selected coded bits are converted into symbols through a modulator (S130) and transmitted via a transmitter (S140).

Meanwhile, in the receiver S150, the transmitted symbol is input, converted into a temporary decision variable in units of coding bits through the demodulator S160, and input to the divider S170. The coded bit distribution and '0' inserter S171 of the distributor divides the temporary decision variable received by the puncturing position calculator S172 and inputs '0' into the punctured position. At this time, the puncturing position calculator (S172) receives the information on the parent code and the puncturing amount through a separate control channel from the code rate controller (S121) of the transmitting end, calculates the puncturing position in real time, and distributes the coded bits. The '0' inserter is notified of the position of the punctured bit, the coded bit distribution and inserter (S171) distributes the received temporary decision variables at positions other than the punctured position, and inserts '0' at the punctured position. do. The temporary decision variables output in this order are input to the decoder S180, decoded, and output as information bits.

Hereinafter, a transceiver for supporting HARQ will be described with reference to FIG. 2.

First, the information bits are input to the encoder S210, converted into encoded bits, and input to the puncturer S220. The encoded bits input to the puncturer are stored in the transmission encoded bit buffer (S221). The encoding bit selector S224 selects and outputs only the encoding bit to be transmitted currently from the buffer S221 according to the information of the puncturing position calculator S223. At this time, the puncturing position calculator (S223) receives the information on the mother code, the puncturing amount, and the number of retransmissions from the HARQ controller (S222) to perform the puncturing position calculation in real time to select the punctured position (S224). The selector selects the coded bits at positions not punctured from the transmit coded bit buffer S221. The output encoded bits are converted into symbols by the modulator S230 and transmitted via the transmitter S240.

The transmitted symbol is input to the receiver S250 and demodulated to a temporary decision variable in units of coding bits in the demodulator S260. The demodulated temporary decision variable is distributed to the received encoded bit buffer S274 through the encoded bit divider S272 according to the information of the puncturing position calculator S271 in the divider S270. At the position of the coded bit that has never been transmitted, '0' is stored in the buffer S274 through the '0' inserter S273 according to the information of the puncturing position calculator. At this time, the puncturing position calculator S271 transmits the punctured position to the encoding bit divider S272 and the '0' inserter S273, and the divider S272 receives the temporary decision variable input to the unpunctured position. It is stored in the corresponding position of the buffer S274. Similarly, the '0' inserter S273 inserts '0' at the punctured position of the reception encoding bit buffer S274. All temporary decision variables including 0 'stored in the buffer S274 are transmitted to the decoder S280, decoded, and passed to the parity checker S280.

The parity checker performs a parity check, and if it passes, transmits an ACK to the transmitting end and detects an information bit. If it fails, the parity checker transmits a NACK to the transmitting end. use.

Hereinafter, a method of puncturing a block-type low density parity check code having a bi-diagonal structure will be described with reference to the accompanying drawings.

여기서 z는 블록의 행 또는 열의 길이이며,

그리고

이다. 블록 쌍 대각 패리티 구조

의 형태는 다음 수학식 1과 같이 표현할 수 있다. In general, the parity check matrix H of a B-LDPC having n columns and m rows can be divided into an information part and a parity part as follows.

Where z is the length of the row or column of the block,

And

to be. Block Pair Diagonal Parity Structure

Can be expressed as Equation 1 below.

는 3개의 0이 아닌 블록을 가지고 있으며

는 쌍 대각 블록을 가진

크기의 행렬이다. 여기서

의 i번째 행과 j번째 열은 i와 j가 같을 때는

크기의 단위행렬 I이며,

일 때는

이다. 여기서

는

크기의 단위행렬을 오른쪽으로

번 순환이동 시키고 나서의 행렬이다.

의 위치는 블록 단위로 l 번째 행에 위치한다. IEEE 802.16e에서 사용되는 B-LDPC부호들은

이고, 효과적인 인코딩을 위해서

, 그리고

으로 설계되어있다. q는 블록 사이즈 z와 서로 소가 되도록 결정된다. 그러면 모든 패리티 블록안의 비트들끼리는

의 길이를 가지는 사이클 한 개와

의 길이를 가지는 사이클 한 개와

의 길이를 가지는 사이클 z개로 서로 연결이 되며, 이를 각각 연결 1, 연결 2 그리고 연결 3 이라 정의한다.Matrix above

Has three non-zero blocks

Has a pair of diagonal blocks

Is a matrix of magnitudes. here

The i th row and the j th column of are equal to i

Unit matrix I of size,

When

to be. here

Is

Unit matrix of size to the right

This is the matrix after the first cycle.

The position of is in the lth row in block units. B-LDPC codes used in IEEE 802.16e

For efficient encoding

, And

Is designed. q is determined to be small with the block size z. Then the bits in every parity block

One cycle with the length of

One cycle with the length of

Z cycles with lengths are connected to each other and are defined as connection 1, connection 2 and connection 3, respectively.

Hereinafter, the above-described block grouping order will be described with reference to FIG. 4.

First, maximize the number of k-SR nodes having the smallest value. Select only blocks of degree 2 here. First, set the (2n + 1) th block as the first block group (n = 0,1, ..., m _b / 2-1). In this way, a maximum number of m _b z / 2 1-SR nodes are obtained, and each 1-SR node updates reliability from all connected check nodes in recovery. If all bits of the first block group are punctured, the 2 (2n + 1) th block is defined as (n = 0,1, ..., m _b / 4-1) as the second block group. In this way, a maximum number of m _b z / 4 2-SR nodes is obtained, and each 2-SR node updates reliability from all connected check nodes in recovery. In this way, in the t-th block group, in the 2 ^t-1 (2n + 1) th block, (n = 0,1, ... m _b -1), the 2 ^t-1 -SR node (t = 1,2 ,. Continue until the number of blocks is maximized and the number of remaining blocks is odd. If the number of blocks remaining is odd, the block with the unpunctured degree 2 becomes the last block group.

Hereinafter, the drilling method will be described with reference to FIGS. 5 and 6.

First, the group index t is 1 and the number i of the current puncturing bits is initialized to 0 (S410). When i becomes Np (S420), the puncturing process is stopped (S490). Np is the total number of bits to puncture. If m _b / 2 ^t is an integer (S430) and there are blocks left unperforated in the current group, i.e. i <(2 ^t -1) m _b z / 2 ^t (S440), the position of the i th puncturing bit Is determined by L ₁ (i) (S450). If all bits in the current block group are punctured (S440), the group index t is increased by 1 (S480). If m _b / 2 ^t is not an integer (S430), that is, immediately after the group index is increased, if the number of remaining blocks is odd, the position of the i th puncturing bit is determined by L ₂ (i). (S460). After punching one bit, i is increased by 1 (S470) and the drilling operation is continued (S420).

The position of the i th puncturing bit may be divided into a block selection and a bit selection step.

As a simple selection method, the leftmost block may be selected from the selected block group and the leftmost bit may be selected from the selected block. In this case, the positioning of the i th puncturing bit is expressed by Equation 2 below.

는

그리고

는 c 보다 작거나 같은 최대 정수를 의미한다. remind

Is

And

Is the largest integer less than or equal to c.

This method is easy and simple, but does not guarantee a uniform distribution of the puncturing bits. Therefore, a bit node having a relatively low reliability is present and performance degradation occurs.

In order to solve the above problem, a uniform selection sequence u _n for uniformly distributing the puncturing bits in the zigzag edge connection is defined as in Equation 3 below.

In the above, k is a natural number and u _n (i), (i = 0,1, ..., n-1) is the i th element in the sequence u _n . The above sequence u _n can be calculated recursively in order to easily obtain in real time.

은 또 다른 균일 선택 수열이 된다. Also the partial sequence of the sequence u _n

Becomes another uniform selection sequence.

Therefore, by using the above uniform selection sequence as the block selection sequence, uniform selection is possible for each connection.

The above uniform selection sequence may also be used for bit selection in order to make uniform even when any bit is punctured in the above-described bit selection in the block.

인 블록들은 내부의 비트 연결을 치환시키게 되므로, 비트 치환정보와 균일 선택 수열을 동시에 사용하여야 지그재그 엣지 연결에서 완전히 균일하게 천공 비트를 할당할 수 있다. 따라서 비트 치환정보를 이용한 변형된 균일 선택 수열은 다음과 같이 수학식 4로 정의한다.However

Since the in-blocks replace the internal bit connection, the bit replacement information and the uniform selection sequence must be used at the same time so that the puncturing bits can be completely uniformly allocated in the zigzag edge connection. Therefore, the modified uniform selection sequence using the bit substitution information is defined by Equation 4 as follows.

Using Equation 4, the determination of the position of the proposed puncturing bit may be expressed as Equation 5 below.

Hereinafter, an embodiment of the present invention will be described with reference to FIG. 7.

The upper part of FIG. 7 shows a case where only the proposed block grouping is used, and the lower part shows an embodiment in which the block selection and the bit selection are uniformly punctured as described above after the block grouping. Where z = 4, b ₁ = q, p = b _i = 0 for 2 <i <8 and the total number of puncture bits is six.

If only block grouping is used, the leftmost unperforated block is selected from the first block group, and the leftmost unperforated bit is selected within that block. In other words, if only block grouping is used, perforation is performed in the order of 5 bit-> 6 bit-> 7 bit-> 8 bit-> 13 bit-> 14 bit. After punching the six bits, it can be seen from the connections 1,2 and 3 that the punched bits are not evenly distributed.

를 사용하여 마지막 2개의 비트는 수열

를 사용하여 위치를 결정하게 된 것이다. However, if block selection and bit selection are punctured by using a uniform selection sequence considering bit substitution, bit 5-> bit 7-> bit 8-> bit 6-> bit 21-> bit 23 Perforations are made in the order of. As a result of the puncturing, it can be seen that the punched bits are uniformly distributed in all three types of connections. The first four bits are a sequence

The last two bits using a sequence

To determine the location.

Through the above results, it can be seen that using the present invention, even in the case of any number of puncturing bits, the puncturing nodes are uniformly distributed in the connections 1,2 and 3.

Hereinafter, the effects of the present invention will be described in detail with reference to FIGS. 8, 9, and 10.

In FIG. 8, the performance of the four puncturing bits is compared based on two code rates. In this case, the B-LDPC code of IEEE 802.16e with code rate 1/2, codeword length 2304 and z = 96 is used as the mother code, and the puncturing code of code rate 5/6 (7/8) is 922 from the mother code. (987) bits were made with the perforation of the present invention.

Since the corresponding puncturing amount is not divided by the size of the block, it can be seen that the code rate as described above cannot be obtained by the conventional puncturing technique.

In FIG. 8, the 'LF' method is a method of puncturing by simply selecting the leftmost bit without using the proposed block group method. In this case, there are many k-SR nodes with large k values, so the performance is very poor.

Secondly, the 'PG / LFB / WBS' method selects the leftmost block after the proposed block group method for block selection, and intentionally selects the right bit after substitution for bit selection (i.e. The worst choice). As can be seen from the performance difference between 'LF' and 'PG / LFB / WBS', the proposed block group method can achieve a significant performance improvement of perforated B-LDPC codes. It is important because it is generally available.

The third grouping method, which uses only the proposed block grouping, is to select the leftmost block and select the rightmost bits in the block after grouping the blocks. The proposed method and the PG / LFB Performance can be achieved between the performance of the / WBS 'method. In addition, there is an advantage that can be used simply without serious performance loss in the absence of substitution information of the parent code.

Fourth, the proposed method of uniformly selecting blocks and uniformly selecting bits in combination with the substitution method in the proposed puncturing bit selection shows the best frame error rate performance as shown in FIG. 6. In addition, it can be seen that the proposed block selection and bit selection have significant performance gains in the performance difference between the proposed method and the PG / LFB / WBS method.

In Fig. 9, a puncturing rate of code rate 1/2 of the IEEE 802.16e code rate, code rate 2/3 obtained from a B-LDPC mother code of codeword length 2304, codeword length 1728, code rate 3/4, and codeword length 1536 are shown. The performance of the B-LDPC code and the uncoded IEEE802.16e code and turbo code with the same code rate and length are compared.

As can be seen in FIG. 9, the perforated code obtained by the present invention is superior in performance except in the environment of a lower signal-to-noise ratio (SNR) than the conventional unperforated code and turbo code.

In Figure 10, code rates 1/3, 2/3, and 3/4, respectively, from code rate 1/3, codeword length 3456 generated B-LDPC mother code under conditions that maximize girth. Three perforated B-LDPC codes of length 2304, 1728, and 1536 were generated by the present invention. And the performance of frame error rate was compared under the same condition as turbo code and unperforated B-LDPC code of IEEE802.16e.

It can be seen from FIG. 10 that the proposed puncturing code outperforms the turbo code and the unpunched code. Therefore, using the puncturing technique of the present invention, a variable code rate B-LDPC code can be obtained from one good mother B-LDPC code, which is superior to the code used in commercial standards.

While the invention has been shown and described with respect to certain preferred embodiments, the invention is not limited to these embodiments, and those of ordinary skill in the art claim the invention as claimed in the appended claims. It includes all the various forms of embodiments that can be implemented without departing from the spirit.

According to the puncturing method of the present invention block type low density parity check code as described above,

First, more efficient bitwise puncturing for a B-LDPC code having a bi-diagonal parity structure can be achieved.

Secondly, it does not need memory to store arbitrary puncturing patterns and can puncture in real time through the formulated puncturing pattern,

Second, it is possible to generate a perforation pattern applicable to the B-LDPC code having a bi-diagonal structure,

Third, it is possible to provide a structural puncturing pattern having the best performance in any bit rate, that is, any code rate change,

Fourth, since the blocks and bits to be punctured are determined according to a predetermined block group, the performance of the punctured code can be basically guaranteed through this.

Fifth, since the perforated bits are uniformly distributed in the zigzag edge connection, the degradation of reliability of the perforated bits and the non-perforated bits adjacent thereto are uniformly distributed to obtain better decoding performance.

Sixth, any form of H-ARQ technique can be applied to the puncture method of the present invention.

Claims (27)

A code rate controller for transmitting information necessary for puncturing; A puncturing position calculator (a) coupled to the code rate controller for calculating a puncturing bit position using given information; A transmitter connected to the puncturing position calculator (a) and including an encoding bit selector for selecting and transmitting an unpunctured bit among the encoding bits input to the puncturer; A puncturing position calculator (b) for receiving information through a control channel to the code rate controller of the transmitter and calculating a puncturing bit position; A receiver connected to the puncturing position calculator (b) and configured to distribute a temporal decision variable in units of coding bits and to insert and output '0' at a puncturing bit position and to output a '0' inserter; Communication system supporting a variable code rate, characterized in that configured to include. The communication system of claim 1, wherein the code rate controller delivers puncturing amount information according to a codeword length, a code rate, and a transmission code rate of a mother code. The method according to claim 1 or 2, The puncturing position calculator (a) of the transmitter and the puncturing position calculator (b) of the receiver calculate a position of a puncturing bit in real time. The method according to claim 3, The puncture position calculator (a), (b), Determining a predetermined block group; Positioning a block within the group; And determining a position of a bit to be punctured in the block. The method according to claim 4, The step of determining the predetermined block group, the communication system supporting a variable code rate, characterized in that the punctured node is recovered with the smallest number of iterations. The method according to claim 5, The step of determining the predetermined block group, the communication system supporting a variable code rate, characterized in that the grouping so that the maximum number of k-SR nodes having the smallest k value. The method according to claim 4, Positioning the block within the group, Checking whether the number of remaining blocks is odd; And determining whether there are blocks left unperforated in the current group. The method according to claim 7, The step of checking whether the number of remaining blocks is an odd number is performed by checking whether m _b / 2 ^t is an integer immediately after t is increased by 1, wherein the value of m _b is the number of rows of blocks, and t is a group index. Communication system that supports a variable code rate, characterized in that for showing. The method according to claim 4, Positioning the bits to be punctured in the block, The number of the remaining blocks the even and to determine the drilling bit positions by, left L ₂ (i) if the number is an odd number of blocks by L ₁ (i) If the block remains without being punctured in the current group, L ₁ (i) and L ₂ (i) are each

And said

Is

And

Is a communication system supporting a variable code rate, the maximum integer being less than or equal to c. The method according to claim 9, L ₁ (i) and L ₂ (i) are each

And u ' _z (i) and u'' _z (i) are each

And u _z (i) is

Communication system supporting a variable code rate, characterized in that. A transmit encoded bit buffer for storing encoded bits; A HARQ controller for controlling feedback signals and information necessary for puncturing; A puncturing position calculator (a) connected to the HARQ controller and receiving the information to calculate the puncturing position; A transmitter connected to the encoding bit buffer and the puncturing position calculator (a) and including an encoding bit selector for selecting only encoding bits to be transmitted; A puncturing position calculator (b) that receives information through a control channel to the HARQ controller of the transmitter and calculates a puncturing position; A coded bit divider connected to the puncturing position calculator (b) to distribute received coded bits; A '0' inserter connected to the puncturing position calculator (b) to insert a '0' into a sub-compensation bit position that has not been transmitted; A receiver coupled to the encoded bit divider and the '0' inserter and including a received encoded bit buffer for storing received encoded bits; Communication system supporting HARQ, characterized in that configured to include. 12. The communication system of claim 11, wherein the HARQ controller transmits puncturing amount information according to a codeword length, a code rate, and a transmission code rate of a mother code. The method according to claim 11 or 12, The puncturing position calculator (a) of the transmitter and the puncturing position calculator (b) of the receiver calculate a position of a puncturing bit in real time. The method according to claim 13, The puncture position calculator (a), (b), Determining a predetermined block group; Positioning a block within the group; And positioning the bits to be punctured in the block. The method according to claim 14, The step of determining the predetermined block group, the communication system for supporting HARQ, characterized in that the punctured node is recovered with the smallest number of iterations. The method according to claim 15, The determining of the predetermined block group may include grouping the maximum number of k-SR nodes having the smallest k value to maximize the number of k-SR nodes. The method according to claim 14, Positioning the block within the group, Checking whether the number of remaining blocks is odd; A communication system supporting HARQ, characterized in that it comprises a step of checking the presence of blocks that are not punctured in the current group. The method according to claim 17, The step of checking whether the number of remaining blocks is an odd number is performed by checking whether m _b / 2 ^t is an integer immediately after t is increased by 1, wherein the mb value is the number of rows of blocks, and t is a group index. Communication system supporting HARQ, characterized in that represented. The method according to claim 14, Positioning the bits to be punctured in the block, The number of the remaining blocks the even and to determine the drilling bit positions by, left L ₂ (i) if the number is an odd number of blocks by L ₁ (i) If the block remains without being punctured in the current group, L ₁ (i) and L ₂ (i) are each

And said

Is

And

Is a communication system supporting HARQ, which means a maximum integer less than or equal to c. The method according to claim 19, L ₁ (i) and L ₂ (i) are each

And u ' _z (i) and u'' _z (i) are each

And u _z (i) is

A communication system supporting HARQ, characterized in that the. In the block-type low density parity check code having a bi-diagonal parity block structure, Determining a predetermined block group; Positioning a block within the group; And determining a position of a bit to be punctured in the block. The method according to claim 21, The step of determining the predetermined block group, the puncturing method of the block-type low density parity check code, characterized in that the punctured node is recovered with the smallest number of iterations. The method according to claim 22, The step of determining the predetermined block group, the method of puncturing block type low density parity check code, characterized in that the grouping so that the number of k-SR node having the smallest k value to the maximum. The method according to claim 21, Positioning the block within the group, Checking whether the number of remaining blocks is odd; A method for puncturing a block type low density parity check code, characterized in that it comprises the step of checking for the existence of blocks that are not punctured in the current group. The method of claim 24, The step of checking whether the number of remaining blocks is an odd number is performed by checking whether m _b / 2 ^t is an integer immediately after t is increased by 1, wherein the mb value is the number of rows of blocks, and t represents a group index. A puncturing method for a block type low density parity check code, characterized in that. The method according to claim 21, Positioning the bits to be punctured in the block, The number of the remaining blocks the even and to determine the drilling bit positions by, left L ₂ (i) if the number is an odd number of blocks by L ₁ (i) If the block remains without being punctured in the current group, L ₁ (i) and L ₂ (i) are each

And said

Is

And

Is a puncturing method of a block type low density parity check code, characterized in that it means a maximum integer less than or equal to c. The method of claim 26, L ₁ (i) and L ₂ (i) are each

And u ' _z (i) and u'' _z (i) are each

And u _z (i) is

A puncturing method for a block-type low density parity check code, characterized in that.

Images