差分算法在盲均衡中的应用与研究  

Application and Research of Differential Evolution in Blind Equalization

作　　者：蔡思捷[1] 于舒娟[1] 张昀[1] 

机构地区：[1]南京邮电大学电子科学与工程学院,江苏南京210003

出　　处：《计算机技术与发展》2012年第6期64-67,共4页Computer Technology and Development

基　　金：国家自然科学基金(60772060)

摘　　要：文中给出了基于差分的盲检测算法,并进行了应用与研究,仿真表明差分盲检测算法具有较强的全局收敛能力和鲁棒性,但是易陷入局部最小值,收敛速度较慢,因此文中对差分盲检测算法做了进一步改进,从而得到改进的差分盲检测算法。改进的差分盲检测算法利用了混沌算法的初值敏感性、拓扑传递性、整体稳定性将其与差分盲检测算法相结合,在算法初始化种群时加入混沌扰动,寻找无规律中的有规律。仿真结果表明了改进算法加快了收敛速度,优化了算法性能。It gives the application and research of differential evolution in blind equalization, simulation shows that differential blind detection algorithm has strong global convergence and robustness,but easy to get into a local minimum and convergence is slow. Blind detection algorithm for differential was further improved in this article,resulting in improved differential blind detection algorithm. Differential blind detection algorithm uses the algorithm of pseudo random chaos, ergodicity,initial sensitivity, topological transitivity, overall stability,combining it with blind detection algorithm,adding chaos disturbance when the population is initialized,looking for irregular regular. Simulation results show that the impoved algorithm accelerates convergence speed and optimizes algorithm performance.

关 键 词：差分盲检测算法 混沌算法 盲均衡 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]