Efficient Decoding of Block Turbo Codes

Abstract

Block turbo codes (BTCs) under iterative decoding are product codes showing excellent performance with reasonable complexity, whose component codes are conventionally decoded in two stages. The Chase algorithm is employed in the first stage to make a list of candidate codewords from the received vector, while the extrinsic information for iterative decoding is generated in the second stage. In this paper, we propose an efficient decoding algorithm for BTCs. The proposed algorithm can avoid a number of unnecessary hard-decision decoding operations by imposing two conditions on the Chase algorithm. Also, it simply computes the extrinsic information for the decision codeword. Numerical results demonstrate that the proposed algorithm has not only much lower computational complexity, but also a little better performance than the conventional decoding scheme based on the Chase algorithm. Furthermore, it can provide a trade-off between the performance and the computational complexity of BTCs by properly selecting a decoding parameter.