基于粒子群算法的跳频信号参数盲估计  

Parameter Estimation of Frequency-Hopping Signal Based on a Modified Particle Swarm Optimizer

作　　者：郭建涛[1,2] 王宏远[1] 

机构地区：[1]华中科技大学电子与信息工程系,湖北武汉430074 [2]信阳师范学院物理与电子工程学院,河南信阳464000

出　　处：《微电子学与计算机》2009年第7期164-167,共4页Microelectronics & Computer

基　　金：国防预研基金项目(51437030105HK101)

摘　　要：提出了一种新的聚类方法,并将其与采用变异操作只有认知模型的微粒群算法相结合,用于跳频信号参数估计.该聚类方法以每个粒子为中心,利用距离测度确定其邻域,在邻域内用适应度最大粒子的标号作为当前粒子标号;然后基于标号对粒子群加以分类.仿真结果表明,基于聚类和变异操作的粒子群算法有较高的解质量及较短的运算时间.This paper presents a new cluster analysis algorithm combined with cognition-only modal mutation particle swarm optimization for solving parameter estimation of frequency hopping signals. This new algorithm takes each particle as a center to determine its neighborhood using distance measure and the particle index whose fitness is biggest within the neighborhood is taken as current particle index. These indices are applied to classify the individuals finally. Simulation results indicate that, to parameter estimation of frequency hopping signal, the modified particle swarm optimization based on cluster and mutation opteration can achieve better search ability and higher convergence speed.

关 键 词：粒子群算法 聚类和变异 跳频信号 参数估计 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]