Piggybacking Belief Propagation Decoding for Rateless Codes Based on RA Structure

Abstract

In this paper, a new Belief Propagation (BP) decoding algorithm named Piggybacking Belief Propagation (PBP) decoding algorithm is proposed for rateless codes based on repeat-accumulate (RA) structure over AWGN channel. The "piggybacking" characteristic of the proposed algorithm is transmitting the Log-Likelihood Ratio (LLR) results calculated from the previous decode attempt to the new decoding process after receiving more information from the channel. Moreover, the proposed algorithm introduces a new stopping criterion to stop the decoding attempt when it is almost impossible to succeed. The stopping criterion is based on the weight change ratio of syndrome. The simulation results show that, compared with the traditional BP decoding algorithm, the proposed algorithm has lower decoding overhead and the a reduced decoding delay.