• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

J4 ›› 2008, Vol. 30 ›› Issue (12): 146-148.

• 论文 • 上一篇    下一篇

恒模盲均衡算法的物理级建模与仿真

刘剑锋 蒋卓勤 盛小平 霍效新   

  • 出版日期:2008-12-01 发布日期:2010-05-19

  • Online:2008-12-01 Published:2010-05-19

摘要:

通过研究恒模盲均衡算法的特点,本文提出了一种基于Simulink的恒模盲均衡算法的建模方法。该方法避开了使用复杂编程语言的实现方式,而是采用基于物理级的动态可视化的建模方法。为了验证仿真模型的正确性和有效性,本文构建了BPSK数字通信系统,并以该通信系统为平台,对其均衡性能进行了仿真分析。为了分析该算法的优缺点,与与基于LMS算法和RLS算法的非盲均衡器进行了对比,给出了三种不同均衡算法的系统误码率、星座图及收敛曲线。仿真结果表明:本文对恒模盲均衡算法的建模方法是正确的,具有很好的对时变信道均衡的性能,并且系统的误码性能很好,而且该算法不需要任何输入信号的先验知识。同时,本文也指出了该算法存在收敛较幔、系统误码率较高的  的缺点。

关键词: 恒模算法 盲均衡 物理级建模 仿真

Abstract:

By studying the characteristics of the constant modulus blind equalization algorithm(CMA), this paper presents a method of modeling CMA based on Sim   ulink. This method avoids using a complex programming language to implement the algorithm, but adopts a visual modeling method based on the physical lay er. In order to verify its validity and effectiveness, a BPSK digital communication system is established. Simulation and analysis is performed on the e qualization performance of the model based on the platform of digital communication, and a comparison is made between LMS, RLS,and CMA. Through the simu   lation, this paper provides the BER performance, constellation map and convergence curve under the condition of three different algorithms. Simulation results show that the model is correct, has good performance to equalize time-varying channels and needs no signal transcendental information. By comparing with the non-blind equalizer based on LMS and RLS, this paper also demonstrates that the convergence of CMA is slower,and BER is higher.

Key words: constant modulus algorithm, blind equalization, physical layer modeling, simulation