Joint Source-Channel Polarization Successive Cancellation Decoding
Ran An, Guojun Liao, Qiufang Lian, Sannyuya Liu, Lin Zhou · 2023
The Double-Polar (D-Polar) code is an innovative scheme that combines source coding and channel coding into a joint coding framework. It aimed at improving the reliability and efficiency of information transmission. By integrating source coding and channel coding, the D-Polar code aims to minimize the bit error rate, enhance the signal’s resilience to interference, and improve its fault tolerance. In this paper, a new type of joint source-channel decoder called Joint Successive Cancellation (J-SC) Decoder is proposed. The proposed approach aims to enhance the decoding performance of D-Polar codes. The merging of nodes representing the information bits of the channel and the nodes corresponding to the high-entropy bits of the source is achieved through the creation of joint source-channel (J) nodes. Based on this joint coding approach, the J-SC decoder is designed to recover the source message by combining source SC decoding with channel SC decoding. During J-SC decoding, the bit values are recovered symbol by symbol. For each J node, the values of source log-likelihood ratios (LLRs) and channel log-likelihood ratios (LLRc) are considered. The simulation results indicate that the decoding performance of J-SC surpasses that of SC decoding, and the performance improves as the overall code rate decreases.