AI-Based Channel Coding for 5G/6G
Madhavsingh Indoonundon, Tulsi Pawan Fowdur, Zoran S. Bojkovic, Dragorad A. Milovanovic · 2023
Machine learning (ML) has been receiving significant research attention in the past few years and has even paved way for a channel coding field in which researchers are actively working to integrate it with conventional channel coding schemes for optimizing the latter’s performance and complexity. Furthermore, with the spotlight currently being on 5G technology, which is expected to bring disruptive changes to mobile connectivity, researchers are focusing to incorporate ML in the two main channel code contenders in 5G, LDPC codes and polar codes, to achieve 5G’s stringent requirements. One such integration of ML in channel coding is the use of an ML-generated factor in each decoding iteration of the LDPC Min-Sum decoder to achieve noticeable block error rate improvements. Another such integration is the use of ML in the polar code Belief-Propagation decoder for selecting good factor-graph permutations during the decoding process to improve the frame error rate. In this chapter, a state-of-the art review of ML techniques that have been, and also that can potentially be, employed to 5G, and 6G channel coding schemes will be performed along with directions for future works.