A Sequence Decoding Method for Large Constrained Length Convolutional Codes

Abstract

A sequential decoding method for a convolutional code with highly-constrained length belongs to the technical field of communication channel encoding and decoding. According to invention, the specific realization process of the sequential decoding of the convolutional code (n, 1, m) is given and the problem of stack overflow which is liable to appear during the decoding process is solved. In fact, the sequential decoding of the convolutional code is a path search process along a code tree according to the norm of smallest measurement valve; path search is free to go forward or back out; paths in different lengths can be obtained; eventually, the path with the greatest measurement valve is taken as a decoding path. In the specific realization process, decoding paths, measurement valves and the next state needed to store in each decoding step are stored through a chain list structure in the C language, so that the programming operation is realized more easily; interval storage and deleting are performed on paths obtained by decoding, so that caches are prevented from increasing unlimitedly. Therefore, the problem of stack overflow is efficiently solved.

Description

technical field [0001] The invention relates to a sequence decoding method for a convolutional code with a large constraint length, and belongs to the technical field of communication channel coding and decoding. Background technique [0002] In the field of channel coding and decoding, convolutional codes are usually used for channel coding to improve the reliability of information transmission. The traditional decoding method is Viterbi decoding, although Viterbi decoding is an optimal decoding, but because the decoding overhead is limited, its decoding calculation will increase exponentially with the constraint length, so Only short constraint length codes can be used. [0003] The sequence decoding algorithm can just solve this problem, and its decoding workload is essentially independent of the constraint length, so a large constraint length can be used. We know that convolutional codes have three geometric representation methods, namely code tree diagram, state diagr...

AI Technical Summary

Problems solved by technology

[0005] In order to overcome the defects and deficiencies in the prior art, the present invention provides a sequence decoding method for large constrained length convolutional codes, so as to achieve a simple and easy data storage method and effectively solve the problem of stack overflow