Efficient Degree Optimization for High-Rate Structured QC-LDPC

In order to efficiently optimize degree distribution of high-rate structured QC-LDPC (SQC-LDPC), based on analysis to the special structure of SQC-LDPC, this paper modifies previous degree distribution optimization method. At first, it proposes a new set of global optimization constraints for Differential Evolution (DE) algorithm. Furthermore, when Gaussian Approximate (GA) is used to compute the threshold of a given degree distribution pair, binary search rather than usually used fixed step search is adopted to speed up degree distribution optimization. Simulation results indicate that proposed degree distribution optimization method can obtain good degree distribution pairs of high-rate SQC-LDPC with less iteration times, and that high-rate SQC-LDPC designed by good degree distribution pairs embraces good performance.