Deep-Polar Codes: An Efficient Error Correction Code for Short Blocklength Transmission

In this paper, we put forth a novel class of pre-transformed polar codes, called as “deep-polar codes.” Encoding of deep-polar codes involves a series of multi-layered polar transformations. By applying these successive polar-based pre-transformations, deep-polar codes with an improved weight distribution are constructed with low encoding complexity. We also propose a low-complexity successive cancellation list decoding algorithm to effectively decode the deep-polar codes. This decoding algorithm harnesses the reverse process of the multi-layered pre-transformed encoding, referred to as “back propagation parity checks.” From simulations, we demonstrate that deep-polar codes outperform existing pre-transformed polar codes in terms of block error rates across various code rates and short blocklengths, while maintaining low-complexity encoding and decoding. Our findings highlight the effectiveness and practicality of deep-polar codes for short blocklength transmission scenarios in 6G.