Adaptive decoding of LDPC codes based on cross-entropy

An adaptive decoding scheme for low-density parity-check (LDPC) codes is proposed for accelerating the convergence of message-passing algorithms. The evolution of adaptive weighting factor with the number of decoding iterations and stopping criterion are based on the cross-entropy between two consecutive a posteriori probabilities distribution of the soft output of decoder. Information-theoretic support and extensive simulations are provided to demonstrate the efficiency of the proposed algorithm for LDPC decoding. The proposed adaptive decoding algorithm can reduce significantly the average number of iterations of decoder while having a comparable or even better performance than conventional message-passing algorithms.