基于ML的多径信道下直扩信号PN码和信道的联合盲估计

Joint Blind Estimation of PN Codes and Channels with Maximum Likelihood for DSSS Signals in Multipath Channels

针对非协作通信下多径信道直接序列扩频信号伪随机(PN,Pseudorandom)码的估计难题,本文在分析信号二阶统计特性的基础上,提出了一种基于最大似然(ML,Maximum Likelihood)的PN码和多径信道联合盲估计的方法.该方法首先建立PN码和信道序列的最大似然数学模型,然后通过交替转换数学模型和使用迭代最小二乘投影(ILSP,Iterative Least Square Projection)算法实现PN码和多径信道的联合估计.为了进一步降低算法复杂度和避免矩阵求逆,本文给出了算法的自适应求解方式.此外,针对低信噪比下信道估计误差引起PN码的估计精度下降的问题,本文提出了一种基于迭代总体最小二乘投影的改进算法.所提算法不受PN码码型限制,并通过仿真实验验证了算法的有效性.

To solve the problem of pseudorandom(PN)codes estimation for direct sequence spread spectrum(DSSS)signals over multipath channels in non-cooperative communication,based on analyzing the second-order statistics of the signals,a method for joint blind estimation of PN codes and channels with maximum likelihood(ML)is proposed.First,we establish a ML mathematical model of PN codes and multipath channels.Then,we iteratively transform the mathematical model and use the iterative least square projection(ILSP)algorithm to estimate the PN code and channel.Furthermore,to reduce the complexity of the algorithm and avoid the matrix inversion,we present an adaptive rule of our algorithm.Finally,to avoid the decrease of PN code estimation accuracy caused by the channel estimation error,especially under low signal-tonoise ratio,an improved algorithm based on the iterative total least squares projection(ITLSP)is presented.The proposed methods are applicable to all types of PN codes and the simulation results are presented to demonstrate the effectiveness of the algorithms.