Low Complexity Syndrome-Based Decoding Algorithm Applied to Block Turbo Codes

doi:10.1109/ACCESS.2018.2829087

Return

Low Complexity Syndrome-Based Decoding Algorithm Applied to Block Turbo Codes

2018-01-01

Byungkyu Ahn

Sungsik Yoon

Jun Heo *

DOI：10.1109/ACCESS.2018.2829087

Original

Save

View PDF

Abstract

En 中文

This paper presents a technique for reducing the decoding complexity of block turbo code with an extended Hamming code as a component code. In conventional decoding algorithms, when an input vector has a zero syndrome, complexity can be reduced by using the hard-input soft-output (HISO) algorithm. Although sufficient error correction can be achieved using hard decision decoding (HDD) of a component code, conventional methods have used the soft-input soft-output (SISO) algorithm for input vectors with a single error. However, when HDD is applied to all input vectors in which the syndrome is detected as a single error, performance loss occurs owing to the occasional presence of input vectors with triple errors. To solve this problem, we used two criteria for distinguishing between instances of single and triple errors. We maximized the applied rates of the HDD-based HISO algorithm depending on whether the criteria were satisfied. The SISO algorithm was applied when the two criteria were not met. In this case, the number of HDD usages can be reduced to half by removing duplicates or unnecessary candidate codewords. Simulation results show that the proposed algorithm can considerably reduce decoding complexity without performance loss compared with conventional algorithms.

Keywords:

Block turbo code (BTC)

soft-input soft-output (SISO)

hard-input soft-output (HISO)

hard decision decoding (HDD)

extended Hamming code

AI Summary

Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.