Bit-flipping for stack decoding of polarization-adjusted convolutional (PAC) codes
Author(s)
Date
2022-09-06Source Title
International Workshop on Signal Design and Its Applications in Communications (IWSDA)
Publisher
Institute of Electrical and Electronics Engineers
Pages
1 - 5
Language
English
Type
Conference PaperItem Usage Stats
2
views
views
1
downloads
downloads
Abstract
While sequential decoding of polarization-adjusted convolutional (PAC) codes constructed using polar rate profile has low computational complexity, their error-correction per-formance falls far short of the theoretical bounds for finite blocklength codes. In this paper, we use the bit-flipping technique in the stack decoding algorithm of PAC codes in order to improve their error-correction performance. This technique maintains the low memory requirements of stack decoding and polar demapper. Additionally, at high SNR values, the number of visits at each level of the decoding tree is almost one. Numerical findings indicate that this approach is capable of outperforming the stack decoding algorithm.
Keywords
Polarization-adjusted convolutional codesPolar codes
Stack decoding
Sequential decoding
Bit flipping