科研之友
微信
新浪微博
Facebook
分享链接
摘要
In this paper, a novel low complexity list decoding algorithm for linear block codes is proposed. The proposed algorithm is based on combining selected columns of the check matrix for a linear block code. By modifying the traditional syndrome decoding algorithm, the proposed list decoding algorithm can achieve a desired error-correcting performance with low decoding complexity. Moreover, the proposed list decoding algorithm not only makes a hard decision codeword from the received soft input value associated with every symbols, but also calculates the soft output value associated with every symbol of the decision codeword. The soft output value is denoted as the reliable metric of the associated symbol of a hard decision codeword. Thus, an elementary soft-insoft-out decoder is constructed. By serial concatenating the elementary soft-in-soft-out decoders, an iterative decoder with any iterations can be established. Since product codes are built with row and column linear block codes, so the iterative decoder can be implemented to decode product codes directly. For the proposed algorithm confines its r-pattern search space to an efficient narrow scope, it has lower complexity while offers good performance compared to the conventional weighting iterative decoding algorithm for product codes. Analyses and simulations demonstrate these advantages of the proposed algorithm.
- 出版日期2002-10
- 单位上海交通大学
