A Low-Complexity Decoder For Turbo Product Codes Based On Extended Hamming Codes

Yaqi Wang,Jun Lin,Zhongfeng Wang

DOI: https://doi.org/10.1109/icct.2018.8599928

2018-01-01

Abstract:In this paper, a modified low-complexity Chasen decoding algorithm for Turbo product codes(TPCs) and the corresponding hardware architecture are proposed. The proposed decoding algorithm is based on a low-complexity Fast Chase algorithm proposed in [1]. To further reduce the computational complexity, the number of the candidate codewords used to generate the extrinsic information is reduced to 2, and a new low-complexity method of calculating the extrinsic information is proposed. The proposed algorithm saves 44% real additions. The numerical simulation results show that the proposed algorithm has negligible performance loss if the component codes of TPCs are hamming codes. Moreover, an efficient hardware architecture for a (128, 120)(2) extended Hamming turbo product code is also proposed. The proposed decoder is synthesised under the TSMC 90-nm CMOS technology. The synthesised results show that the proposed decoder reaches a 10.5 Gbps throughput with an area of 1.7 Mgates.