Flooding and node-wise RBP sequentially concatenated decoder for LDPC codes

A new strategy of belief propagation (BP) decoder for LDPC codes is presented by sequentially concatenating the scheduling algorithms of flooding and node-wise residual belief propagation (NW RBP) without increasing the complexity of hardware implementation. Compared to NW RBP, the concatenated BP decoder significantly decreases the decoding latency caused by computing and sorting the residual of each message in the factor graph of an LDPC code. As the signal-to-noise ratio (SNR) increases, the decoding performance of the concatenated BP decoder outperforms that of purely applying flooding or NW RBP, and the average number of decoding iterations required for the concatenated decoder decreases and tends to be approximately the same as the other two.