为解决低信噪比下长伪码序列快速捕获问题,提出一种基于迭代消息传递算法的快速捕获方法.通过证明m序列是一种特殊的线性分组码,将m序列用Tanner型因子图表示,然后在得到的图模型上迭代运行消息传递算法来估计m序列的初始相位.在低信噪比下的分析和仿真结果均表明:该算法能稳定工作.相比混合捕获算法,其检测性能下降了大约5dB,但其平均捕获时间约为同条件下混合捕获方法的1/10,其计算复杂度仅为传统捕获算法的几十分之一.
A rapid acquisition based on iterative message passing algorithm (iMPA) was developed to solve the problem of rapid acquisition for a long pseudorandom (PN) code under low SNR. Tanners' factor graph was u~d to represent the m sequence via demonstrating that m sequence is a sort of especial linear block code, and then the message passing algorithm was executed iter- atively to estimate the original phase of the received m sequence over the graph model. Both analysis and simulation results show that the rapid acquisition using iMPA can work steadily under low SNR, and its detection performance drops about 5 dB com- paring with hybrid acquisition. But the average acquisition time is about one tenth of the hybrid acquisition in the same acquisi- tion environment, and the computing complexity is only one over tens of the traditional acquisition.