Research On High Performance Error Correcting Codes With Iterative Decoding

The future ultra high-speed communication systems demand channel codingschemes with high performance, including high throughput and ultra low error bit rate.Low-density parity-check (LDPC) codes can be decoded with parallel iterativedecoding algorithms and applied in communication systems with high throughput. Inaddition, LDPC codes with large code length can approach the Shannon limit.However, due to error floor phenomena at relatively high signal-to-noise ratios (SNR),channel coding schemes with LDPC codes can not satisfy the requirement for ultralow error bit rate. Therefore, high-performance error correcting codes based on LDPCcodes, which can provide high throughput and low bit error rate, are proposed in thisthesis.On one hand, for product codes based on LDPC codes and BCH codes, aniterative decoding scheme with low complexity is presented. In this scheme, the errorbits corrected and the correct BCH codes are marked in the decoding process of BCHcodes. Then the prior information of LDPC codes is updated by the information ofBCH codes in the next iteration. Simulation results show that, the proposed iterativedecoding scheme outperforms the scheme without iteration and lowers the error floorefficiently.On the other hand, to improve the performance of channel coding scheme indigital terrestrial multimedia broadcasting (DTMB), high performance product codes,which is based on LDPC codes and BCH codes, are proposed. The presented productcodes can lower the error floor of LDPC codes and outperform the concatenated codeswith comparative complexity. Simulation results show that, compared with theconcatenated codes in DTMB, the proposed product codes can achieve a little codinggain.