Critical-Set-Aided Multi-Bit Successive Cancellation List Decoders for Polar Codes

The utilization of 5G and beyond wireless communication systems in autonomous driving and remote healthcare has necessitated the development of channel codes with high reliability and low latency. The multi-bit successive cancellation list (MSCL) parallel decoding of polar codes has shown great potential in the decrease of decoding latency while maintaining a high level of reliability, in comparison to the conventional successive cancellation list (SCL) decoding. However, the complexity of decoding path sorting increases exponentially with the increase of parallelism, which poses significant limitations to the improvement of parallelism. To address this issue, this paper proposes a critical set aided multi-bit successive cancellation list (CS-MSCL) decoding algorithm. Critical sets are introduced to the MSCL decoding, which prunes less reliable candidate paths to guarantee high decoding performance. Meanwhile, dynamic configuration is allowed according to the application scenario to make a good balance in decoding latency, error correction performance, and algorithmic complexity. Simulation results demonstrate that the algorithmic complexity decreased by 1099.04% compared to MSCL decoding for (1024, 512) polar codes when the degree of parallelism is 8 and the list size is 16.