基于改进最大期望算法迭代相位检测器的研究

近年来出现的基于最大期望(EM)迭代算法的相位同步算法具有比传统非数据辅助(NDA)相位估计算法更低的相位误差,是适合于Turbo编码系统的一种相位同步算法。然而,EM迭代算法对初值的选取比较敏感,收敛性较差。为了克服EM迭代算法收敛性较差的缺点,提出一种基于改进EM算法的Turbo编码系统迭代相位同步算法。针对EM算法的缺点,用数据辅助(DA)和NDA结合的方法得到的相位估计作为EM算法的初值。数值仿真显示,该方法具有比随机初始化EM算法更好的相位检测性能。

Study of Iterative Phase Synchronizer Based on Modified EM Algorithms

The Expectation-Maximum(EM) algorithm based phase synchronization scheme proposed recently behaves better in phase tracking performance than conventional Non-Data-Aided(NDA) algorithm,and suit for turbocoded systems.Whereas,the EM algorithm has a bad convergence performance because of its sensitivity to initial values.In order to improve the bad convergence of the EM algorithm,an iterative phase synchronization algorithm based on modified EM algorithm for turbo coded systems is proposed in this paper.The problem is solved by a suited initial value derived from the combined Data-Aided(DA) algorithm and NDA algorithm.Number simulation results show that the proposed scheme has a better phase tracking performance then the random-initial-value EM phase detecting algorithm.