阅读说明：本技术 一种超低码率内部交织卷积编码译码方法 (Ultra-low code rate internal interleaving convolutional coding and decoding method ) 是由 黎光洁 王明威 李春宏 吴冶 于 2018-08-07 设计创作，主要内容包括：本发明公开了作为一种超低码率内部交织卷积编码方法，包括(1)将字节转换为组；(2)将组依次输入移位寄存器；(3)抽取移位寄存器抽头，利用抽头值计算校验组；(4)生成最后编码字节流，所述编码字节流中构成组的信息字节排列在前，构成校验组的校验字节排在所述信息字节之后。内部交织卷积编码码率超低，复杂度低，接近shanoon极限，提供了充分的灵活性，可为高码率和超低码率(1/16?1/30)场景提供优良的性能。(The invention discloses an inner interweaving convolution coding method with ultra-low code rate, which comprises the steps of (1) converting bytes into groups; (2) sequentially inputting the groups into a shift register; (3) extracting a tap of the shift register, and calculating a check group by using a tap value; (4) generating a final coded byte stream, wherein information bytes forming a group in the coded byte stream are arranged in front, and check bytes forming a check group are arranged behind the information bytes. The code rate of the inner interleaving convolutional coding is ultralow, the complexity is low, the inner interleaving convolutional coding is close to the shanoon limit, sufficient flexibility is provided, and excellent performance can be provided for scenes with high code rate and ultralow code rate (1/16-1/30).)

1. An ultra-low bit rate inner interleaving convolutional coding method, comprising:

(1) converting bytes into groups;

(2) sequentially inputting the groups into a shift register;

(3) extracting a tap of the shift register, and calculating a check group by using a tap value;

(4) generating a final coded byte stream, wherein information bytes forming a group in the coded byte stream are arranged in front, and check bytes forming a check group are arranged behind the information bytes.

2. The ultra-low bit rate inner interleaving convolutional coding method of claim 1, wherein the taps are groups extracted from a shift register, the tap values are values of the groups extracted from the shift register, and the number and positions of the groups extracted are determined by system configuration parameters.

3. The ultra-low bit rate inner interleaving convolutional coding method as claimed in claim 1, wherein the method for calculating the parity groups by using the tap values comprises:

(3-1) carrying out non-binary convolution operation on the tap values to obtain a first check group;

(3-2) interleaving the taps according to bytes to generate a new group, and performing non-binary convolution operation on the values of the new group to obtain a second check group;

and (3-3) interleaving the groups in the step (3-2) according to bytes to generate new groups, carrying out non-binary convolution operation on the values of the new groups to obtain the next check group, and repeating the steps until the whole check group is obtained.

4. The ultra-low bit rate inner interleaving convolutional coding method of claim 3, wherein a new group is generated by byte interleaving for the group by using a local interleaver.

5. The ultra-low bit rate inner interleaving convolutional coding method of claim 4, wherein the byte address of the new group generated by byte interleaving can be obtained by table lookup.

6. The ultra-low bit rate inner interleaving convolutional coding method of claim 1, wherein when the number of the parity check groups is k, the coding rate is 1/(1+ k).

7. The method is characterized in that the decoding process is to carry out non-binary iterative decoding on a check group to obtain a group.

8. The method of claim 7, wherein when the parity check groups are subjected to non-binary iterative decoding, the number of groups obtained by each decoding is the same.

9. The ultra-low bit rate inner interleaving convolutional decoding method of claim 7, wherein the starting groups of the groups obtained by the adjacent decoding are adjacent, and the starting group of the group obtained by the previous decoding is previous.

10. The ultra-low bit rate inner interleaving convolutional decoding method of claim 7, wherein the sequence numbers of the groups obtained by the same decoding are separated by one bit.

Technical Field

The invention relates to channel coding, in particular to an ultra-low code rate internal interleaving convolutional coding and decoding method.

Background

Channel coding is to add a certain number of redundant symbols in an information code to make them satisfy a certain constraint relation, and the information symbols and parity symbols constitute a code word transmitted by a channel. Once transmission errors caused by interference of a physical medium and unavoidable noise occur, the constraint relationship between the information code elements and the supervision code elements is destroyed, and the constraint relationship is checked at a receiving end according to a set rule, so that the purpose of finding and correcting the errors is achieved.

Channel coding is the most important part for obtaining gain in a communication system, the convolution of the Turbo code and the code of the traditional channel code such as LTE is designed for the signal-to-noise ratio above 0dB, the code rate is 1/3 at the minimum, the lower code rate needs to be obtained by repetition, and the code has a great distance from the shanoon limit when the code length is shorter.

Disclosure of Invention

Based on the defects in the prior art, the invention provides an internal interleaving convolutional coding and decoding method which is ultra-low in code rate, low in complexity, close to the shanoon limit and flexible in application.

The technical scheme adopted by the embodiment of the invention is as follows:

as an ultra-low bit rate inner interleaving convolutional coding method, one embodiment of the present invention includes:

(1) converting bytes into groups;

(2) sequentially inputting the groups into a shift register;

(3) extracting a tap of the shift register, and calculating a check group by using a tap value;

(4) generating a final coded byte stream, wherein information bytes forming a group in the coded byte stream are arranged in front, and check bytes forming a check group are arranged behind the information bytes.

Further, the taps are groups extracted from the shift register, the tap values are values of the groups extracted from the shift register, and the number and positions of the extracted groups are determined by system configuration parameters.

Further, the method for calculating the check group by using the tap value comprises the following steps:

(3-1) carrying out non-binary convolution operation on the tap values to obtain a first check group;

(3-2) interleaving the taps according to bytes to generate a new group, and performing non-binary convolution operation on the values of the new group to obtain a second check group;

and (3-3) interleaving the groups in the step (3-2) according to bytes to generate new groups, carrying out non-binary convolution operation on the values of the new groups to obtain the next check group, and repeating the steps until the whole check group is obtained.

Further, a new group is generated by byte interleaving the group with a local interleaver.

Further, the byte addresses of the new groups generated by byte interleaving may be obtained by table lookup.

Furthermore, when the number of check groups is k, the coding rate is 1/(1+ k).

As an ultra-low bit rate internal interleaving convolutional decoding method, one embodiment of the present invention is to perform non-binary iterative decoding on a parity check group to obtain a group.

Further, when the check group is subjected to non-binary iterative decoding, the number of groups obtained by decoding each time is the same. Further, the starting group of the group obtained by the adjacent decoding is adjacent, and the starting group of the group obtained by the previous decoding is previous.

Further, the sequence numbers of the groups obtained by the same decoding are separated by one bit.

The positive beneficial technical effects which can be realized by the embodiment of the invention comprise: the code rate of the internal interleaving convolutional coding is ultralow, the complexity is low, and the code rate is close to the shanoon limit; sufficient flexibility is provided, and excellent performance can be provided for high code rate and ultra-low code rate (1/16-1/30) scenes; the local interleaving complexity is low, the implementation is easy, the non-binary iterative decoding can avoid complex interleaving processing, and the complexity is low.

Other aspects and advantages of the present invention will become apparent from the following detailed description, taken in conjunction with the accompanying drawings, illustrating by way of example the principles of the invention.

Drawings

The invention will now be described, by way of example, with reference to the accompanying drawings, in which:

FIG. 1 is a diagram of the relationship between groups and bytes provided by an embodiment of the present invention;

fig. 2 is a schematic diagram of an ultra-low code rate inner interleaving convolutional coding process provided in the embodiment of the present invention.

FIG. 3 is a diagram illustrating a final structure of a code byte stream according to an embodiment of the present invention;

FIG. 4 is a schematic view of an iterative decoding process according to an embodiment of the present invention;

FIG. 5 is a schematic diagram of a non-binary iterative decoding process performed on a parity check group according to an embodiment of the present invention;

Detailed Description

All of the features disclosed in this specification, or all of the steps in any method or process so disclosed, may be combined in any combination, except combinations where mutually exclusive features and/or steps are present.

Any feature disclosed in this specification may be replaced by alternative features serving equivalent or similar purposes, unless expressly stated otherwise. That is, unless expressly stated otherwise, each feature is only an example of a generic series of equivalent or similar features.

An ultra-low code rate internal interleaving convolutional coding method comprises the following steps:

(1) converting bytes into groups;

(2) sequentially inputting the groups into a shift register;

(3) extracting a tap of the shift register, and calculating a check group by using a tap value;

(4) generating a final coded byte stream, wherein information bytes forming a group in the coded byte stream are arranged in front, and check bytes forming a check group are arranged behind the information bytes.

As an example, as shown in fig. 1, one group (group) is formed of 4 bytes (bit), bytes bit0, bit1, bit2, and bit3 are formed into a group0, bytes bit4, bit5, bit6, and bit7 are formed into a group1, and so on, n bits are formed into n/4 groups.

The groups are sequentially input into a shift register, and the initial value of the shift register is the value of the last groups. Each time a shift is made, a group is extracted from the shift register as a tap, the value of the extracted group is a tap value, and the number and position of the extracted groups are determined by system configuration parameters.

The method for calculating the check group by using the tap value comprises the following steps:

(3-1) carrying out non-binary convolution operation on the tap values to obtain a first check group;

(3-2) interleaving the taps according to bytes to generate a new group, and performing non-binary convolution operation on the value of the new group to obtain a second check group;

and (3-3) interleaving the groups in the step (3-2) according to bytes to generate new groups, carrying out non-binary convolution operation on the values of the new groups to obtain the next check group, and repeating the steps until the whole check group is obtained.

As an example, as shown in FIG. 2, group g is extracted from the shift register_a、g_b、g_c、g_dAs a tap, for g_a、g_b、g_c、g_dCarrying out non-binary convolution operation on the value to obtain a first check group p _ g 0; for g_a、g_b、g_c、g_dInterleaving according to bytes to generate a new group, and carrying out non-binary convolution operation on the value of the new group to obtain a second check group p _ g 1; and repeating the interleaving of the groups in the previous step according to bytes to generate new groups, carrying out non-binary convolution operation on the values of the new groups to obtain the next check group, and repeating the operation until k check groups are obtained. Non-binary convolution operation and byte-wise interleaving are both conventional operations and are not described in any further detail herein.

As shown in FIG. 3, the last encoded byte stream consists of g, p _ g0, p _ g1, p _ g2, …, and p _ gk-1, with the information bytes that make up group g arranged in front and the check bytes that make up check group p _ g0, p _ g1, p _ g2, …, and p _ gk-1 arranged in the back. And when the number of the check groups is k, the coding code rate is 1/(1+ k).

The interleaver is the key of the channel coding performance, and a general interleaver performs global interleaving, that is, only one interleaver is used for interleaving all bytes, so that the interleaver is high in complexity and not beneficial to implementation. The method and the device utilize the local interleavers to interleave the groups according to the bytes to generate new groups, the number of the local interleavers is more than one, complexity of encoding and decoding can be greatly reduced, constraint depth can be improved in a multiplied mode, and encoding performance is improved. When k groups of check code streams need to be generated, at least k local interleavers are needed, the local interleavers are interleaved by taking bytes as units, and the interleaved byte addresses can be obtained by table lookup. As an embodiment, the number of taps is 4, the length of the non-binary group is 4 bytes, the interleaving process is performed by interleaving in units of bytes, the number of taps is 4, the byte addresses after 16-byte interleaving are obtained by querying the following table, corresponding to 4 groups and 16 bytes.

An ultra-low code rate internal interleaving convolution decoding method is characterized in that a decoding process is to perform non-binary iterative decoding on a check group to obtain a group, and an iterative decoding flow is shown in fig. 4. When the check group is subjected to non-binary iterative decoding, the number of groups obtained by decoding each time is the same; the initial group of the group obtained by the adjacent decoding is adjacent, and the initial group of the group obtained by the previous decoding is before; the sequence numbers of the groups obtained by the same decoding are separated by one bit. The invention provides an advanced check node (A _ C) concept, which is different from the traditional Belif-propagation (BP) algorithm, and the advanced check node needs to perform local iterative decoding once besides using a non-binary check decoding algorithm. Inside the A _ C, iterative operation is carried out among different check groups, external information (needing interleaving processing) is exchanged among each iterative element, and non-binary check byte processing can adopt trellis min-sum and other non-binary decoding algorithms.

As an embodiment, as shown in FIG. 5, the parity group p _ g0, p _ g1, p _ g2, …, p _ gk-1 are subjected to non-binary iterative decoding (A _ C), and the first decoding obtains the group g₀、g₂、 g₄Second decoding to obtain group g₁、g₃、g₅Third decoding to obtain group g₂、g₄、g₆…, the number of groups obtained by each decoding is 3, and the start groups of the groups obtained by the adjacent decoding are adjacent (the start group g of the groups obtained by the first and second decoding)₀、g₁Adjacent), the starting set of sets obtained from the previous decoding is preceded by (g)₀In g₁Previously), the serial numbers of the groups obtained by the same decoding are separated by one bit (group g obtained by the first decoding)₀、g₂、 g₄One bit apart).

The different aspects, embodiments, implementations or features of the invention can be used alone or in any combination.

The invention is not limited to the foregoing embodiments. The invention extends to any novel feature or any novel combination of features disclosed in this specification and any novel method or process steps or any novel combination of features disclosed.