Constructional Reaearch Of Advanced Channel Code Technology

LDPC (Low Density Parity Check) codes are one of the hottest topics in coding theory today, which approach the Shannon Limit. The main aspects of the LDPC codes are the most important achievement in modern coding theory, such as the iterative decoding algorithm:Message Passing Algorithm, the performance analysis tools:Density Evolution and EXIT Chart, description and design methods based on Tanner Graph etc,LDPC codes exhibit the threshold phenomenon:as the bloke size goes to infinite, an arbitrarily small bit error rate(BER) can be achieved when the noise level is lower than a certain threshold. Richardson and Urbanke developed the density evolution algorithm to determine the thresholds of the LDPC codes ensembles defined by their degree distributions. Besides, an optimization method has recently been presented by Brink to design the degree distribution of irregular LDPC codes using EXIT chart, which is easier to visualize and program compared to density evolution. This paper optimizes the degree distribution of irregular LDPC codes by means of density evolution and EXIT chart combining differential evolution which is a powerful mathematical optimization algorithm, and construct LDPC codes using PEG algorithm. Simulation results are given in terms of the bit error rate under message passing algorithm. Although there is difference between the actual noise level and the theory value, the performance of constructed LDPC codes is pretty good. Inspired by the research of LDPC codes, we propose an algorithm, called CPEG, to construct Tanner graph representing the constraints of chips in PN sequence, which makes the iterative acquisition algorithm applicable for general PN sequences. The resulting iterative acquisition method is much better than the existing iterative method: iMPA both in terms of acquisition performance and complexity. Simulation results show that the performance of iterative acquisition using CPEG graph is improved by 2.5dB, and the complexity is reduced to the half. In order to further improve the acquisition performance, we proposed an iterative acquisition method based on graph-concatenation. Simulation results show that the graph-concatenation method can get 0.3dB processing gain with the same complexity.