摘要
为了获取目标位置的估计值，目前的计算方法主要有迭代法和网格搜索法。迭代法的主要缺点是迭代的初始位置较难确定，网格搜索法的主要缺点是计算量大。以最大似然估计（MLE）定位法为例，针对网格搜索法计算量大的问题，分别用两步网格搜索法和粒子群优化（PSO）搜索法寻求最大似然估计定位的全局最优解，显著减小了计算量。仿真实验以测向交叉定位法为例，说明两步网格搜索法和PSO搜索法计算量小，且PSO搜索法定位精度较高，通过搜索法得到的定位结果可以作为迭代法迭代的初始位置。
Iterative method and grid search method are the main methods to estimate the initial position of target. Main disadvantage of iterative method is difficult to decide initial position, and grid search method is large in computing capacity. In order to solve the problem of large computation of grid search method, two steps grid search method and Particle Swarm Optimization （PSO） search method are used to figure out the global optimal solution of Maximum Likeli- hood Estimation（MLE） location, which significantly reduce the amount of computation. Simulation experiments take bearing-crossing locating as an example, the results show that the two methods can reduce the amount of computation significantly and PSO algorithm has higher location accuracy. The position of grid search method can be used as the initial position in iterative method.
出处
《电子对抗》
2014年第3期16-21,共6页
Electronic Warfare
关键词
网络搜索法
PSO算法
迭代法
初始位置
grid search method
Particle Swarm Optimization （PSO） algorithm
iterative method
initial position
