Generalized Mutual Information-Maximizing Quantized Decoding of LDPC Codes With Layered Scheduling

In this paper, we propose a framework of the mutual information-maximizing (MIM) quantized decoding for low-density parity-check (LDPC) codes by using simple mappings and fixed-point additions. Our method is generic in sense that it can be applied to LDPC with arbitrary degree distributions, implemented based on either belief propagation (BP) algorithm or min-sum (MS) algorithm. particular, MIM density evolution (MIM-DE) construct lookup tables (LUTs) node updates. The computational complexity implementation are discussed compared LUT decoder variants. To accelerate convergence speed quasi-cyclic codes, consider layered schedule, develop MIM-DE design LUTs MS algorithm, leading (MIM-LQMS) decoder. An optimization further introduced reduce memory requirement storing LUTs. Simulation results show decoders outperform state-of-the-art waterfall region both 3-bit 4-bit precision. Moreover, MIM-LQMS approach error performance floating-point BP within 0:1 dB.