ADMM-based Decoder for Binary Linear Codes Aided by Deep Learning
This work addresses decoding efficiency for communication systems, but it is incremental as it builds on existing ADMM and deep learning methods.
The authors tackled the problem of decoding binary linear codes by developing a deep learning-aided algorithm based on unfolding the ADMM-penalized decoder, resulting in improved performance over the original decoder for various LDPC codes with similar computational complexity.
Inspired by the recent advances in deep learning (DL), this work presents a deep neural network aided decoding algorithm for binary linear codes. Based on the concept of deep unfolding, we design a decoding network by unfolding the alternating direction method of multipliers (ADMM)-penalized decoder. In addition, we propose two improved versions of the proposed network. The first one transforms the penalty parameter into a set of iteration-dependent ones, and the second one adopts a specially designed penalty function, which is based on a piecewise linear function with adjustable slopes. Numerical results show that the resulting DL-aided decoders outperform the original ADMM-penalized decoder for various low density parity check (LDPC) codes with similar computational complexity.