粒子滤波重采样及在盲均衡中的应用

提出了一种新的粒子滤波重采样算法——降序二分法,这种算法的主要思想是在分层重采样算法的基础上,寻找权重最大点的过程用折半二分法。另外,把这种算法应用于信道的盲均衡中,为了易于采样,重要性函数采用先验概率密度,用信道的均值代替信道的真实值,仿真结果表明,这种算法的平均性能优于之前的重采样算法,且均衡的同时能够完成对信道的辨识。

Application of Particle Filter Resampling in Blind Equalization

A novel resampling algorithm named Descending Dichotomy Algorithm is proposed in this paper.Based on the stratified resampling algorithm,Dichotomy Algorithm is used in the searching process of the biggest weight pot.In addition,the Descending Dichotomy Algorithm is applied to the blind equalization,and the importance sampling process could be simplified by choosing the prior density as the importance function.Also,for the same reason,the actual value of the channel is replaced by the average value.The simulation results reveal that the average performance of the Descending Dichotomy Algorithm is superior to that of the other resampling algorithms,and at the same time of equalization,the blind identification could be accomplished.