Genetic Algorithm-Optimized Adaptive Offset Min-Sum LDPC Decoder for 5G- NR Base Graph 2 Codes

This paper proposes a resource-efficient Adaptive Offset Min-Sum (A-OMS) LDPC decoder for 5G-NR Base Graph 2 (BG2) codes. A Genetic Algorithm (GA) is employed to optimize the offset factor within the check-node update, addressing the difficulty of selecting a globally optimal offset for the irregular BG2 structure. The GA operates as a meta-heuristic optimizer that minimizes the bit-error rate (BER) over a wide range of signal-to-noise ratios (SNRs). Simulation results show that the proposed GA-optimized A-OMS decoder achieves performance within 0.2–0.4 dB of belief-propagation decoding, while providing coding gains of 0.3–0.6 dB and up to a 4.6× BER reduction compared with conventional fixed-offset OMS. These improvements are obtained with approximately 35% of the computational complexity of BP decoding. The proposed approach offers a practical balance between performance and implementation efficiency for next-generation 5G-NR receivers.