非线性信道的均衡算法研究  

Equalization Algorithms Based on Nonlinear Channel

作　　者：刘顺兰[1] 蒋树南[1] 

机构地区：[1]杭州电子科技大学通信工程学院,浙江杭州310018

出　　处：《电子学报》2010年第10期2219-2223,共5页Acta Electronica Sinica

基　　金：国家级重点实验室基金(No.060406016)

摘　　要：本文使用Hammerstein模型和维纳模型代替Volterra级数模型来模拟非线性结构以降低运算复杂度,提出了一个由Hammerstein模型和维纳模型构建成的非线性信道传输系统的模型.基于该系统模型,分别提出并推导了三种非线性信道的均衡算法:NCRLS算法、NCKalman算法和NCRPEM算法,并对这三种新算法的性能进行了比较.仿真结果表明,在剩余均方误差方面三种算法中NCKalman算法最小,NCRPEM算法次之,NCRLS算法较差;在收敛速度方面NCRPEM算法收敛最快,NCRLS算法次之,NCKalman算法较差.In order to reduce the computational complexity,this paper adopts Hammerstein model and Wiener model,instead of the Volterra series model,to approach a non-linear system.Furthermore,based on Hammerstein model and Wiener model,a nonlinear transmission system model is also proposed.To this new model,three non-linear algorithms,i.e.Nonlinear Channel Recursive Least Squares（NCRLS） algorithm,Nonlinear Channel Kalman（NCKalman） algorithm and Nonlinear Channel Recursive Prediction Error Method（NCRPEM） algorithm,are derived.Also comparisons on performance among these three algorithms are presented.Results of simulations show that,if residual mean square error is concerned,NCKalman performs over NCRPEM and NCRLS is the worst,while NCRPEM is the best,and NCRLS performs over NCKalman if the speed of iterative convergence is considered.

关 键 词：非线性信道 HAMMERSTEIN模型 维纳模型 RLS算法 KALMAN算法 RPEM算法 

分 类 号：TN911.5[电子电信—通信与信息系统]