Deep Learning Methods for Improved Decoding of Linear Codes
This work addresses the challenge of efficient decoding in communication systems, offering incremental improvements to existing methods for researchers and engineers in coding theory.
The paper tackles the problem of low-complexity, near-optimal channel decoding for linear codes with short to moderate block lengths by using deep learning methods to enhance standard belief propagation and min-sum decoders, achieving comparable results with fewer parameters through recurrent neural network architectures and showing improvements on sparser Tanner graphs and for BCH codes.
The problem of low complexity, close to optimal, channel decoding of linear codes with short to moderate block length is considered. It is shown that deep learning methods can be used to improve a standard belief propagation decoder, despite the large example space. Similar improvements are obtained for the min-sum algorithm. It is also shown that tying the parameters of the decoders across iterations, so as to form a recurrent neural network architecture, can be implemented with comparable results. The advantage is that significantly less parameters are required. We also introduce a recurrent neural decoder architecture based on the method of successive relaxation. Improvements over standard belief propagation are also observed on sparser Tanner graph representations of the codes. Furthermore, we demonstrate that the neural belief propagation decoder can be used to improve the performance, or alternatively reduce the computational complexity, of a close to optimal decoder of short BCH codes.