Speed matching method of limited length circular buffer of LDPC code

Abstract

The invention discloses a speed matching method of the limited-length circular cache of a low-density parity-check code. The method includes the following steps of conducting a structured low-density parity-check code encoding on the data bit of an inputted information block, and determining the size of the one-dimensional limited-length circular cache according to the encoding result; selecting a redundant version value from a plurality of preset redundant version values according to the retransmission times of the mixed retransmission request, and determining the initial position of the data bit of the formed mixed retransmission request data packet read from the one-dimensional limited-length circular cache according to the selected redundant version value; and reading the mixed automatic retransmission request data packet composed of specific-length data from the initial position in order, and sending the mixed automatic retransmission request data packet out. The respective corresponding initial positions of the multiple preset redundant version values are evenly or almost evenly distributed on the one-dimensional limited-length circular cache.

Description

technical field [0001] The present invention relates to the communication field, and more specifically relates to a rate matching method of a limited-length circular buffer of a low-density parity-check code. Background technique [0002] Low Density Parity Check Code (LDPC code for short) is a linear block code based on a sparse parity check matrix. It is precisely by using the sparsity of its check matrix that low-complexity encoding and decoding can be achieved. . Structured LDPC codes are a special kind of LDPC codes due to their structural features. They can be coded with variable code length by extension and correction, and codes with variable code rate can be realized by puncturing and expansion. [0003] Define the parity check matrix of the structured LDPC code as a (m b ×z)×(n b ×z) matrix H, which consists of cyclically shifted identity matrix blocks or zero matrices of size z×z, in the form: [0004] H = ...

AI Technical Summary

Problems solved by technology

[0013] In the existing rate matching method of finite-length circular buffer, the HARQ starting points corresponding to each redundant version are unevenly distributed in the circular buffer, which may cause some codewords to overlap multiple times during the HARQ packet retransmission process. The other part of the codewords does not overlap, which will lead to a decline in retransmission performance, and the existing technology does not better combine the characteristics of LDPC structured coding, and cannot give full play to its best retransmission performance

Images