An Improved Segmented Flipped Successive Cancellation List Decoder for Polar Codes
Yunlong Peng, Xingcheng Liu, Jingyun Bao · 2020
Several studies have documented that segmentation can improve the performance of polar codes based on CRC-aided successive cancellation list (CA-SCL) decoding algorithm. In this paper, we propose a new segmented CA-SCL decoder, referred to as improved segmented flipped successive cancellation list (ISF-SCL) decoder, with a reasonable segmentation strategy and a number of extra flipping attempts. Different from the existing methods, in which the uniform segmentation or LLR based segmentation scheme was adopted, a novel segmentation strategy that we firstly analyzed and proposed is based on the distribution of the correct path elimination. Hence, the segment effectiveness is promoted. In addition, to further improve the error correcting performance, a multi-path flipped strategy is proposed to restore the correct path eliminated from the list. Simulation results show that the performance of the ISF-SCL decoding algorithm is superior to the existing segmented CA-SCL decoding algorithms in the low signal-to-noise ratio regime, while the average computational complexity does not increase significantly.