LLR Distribution Characteristics Based Hybrid Decoding Algorithm for LDPC Codes

To improve the decoding speed and reduce the complexity of low-density parity-check (LDPC) codes, this paper proposes a hybrid LDPC decoding algorithm that combined with the distribution characteristics of log-likelihood ratio (LLR) of all variable node information before each iteration. During the process of decoding, there is a judgment based on an interval value and a threshold value to select the Belief Propagation (BP) algorithm or the Minimal Sum (MS) algorithm to decode, so the proposed hybrid algorithm effectively combines the excellent performance of BP algorithm and the simple operation of MS algorithm. The simulation results show that the algorithm can greatly reduce the computational complexity and achieve similar decoding performance compared with the BP algorithm.