摘要
利用最小二乘法将超声回波参数估计问题转化为优化问题,采用改进的差分进化算法(DE算法)对待优化函数进行优化处理。仿真研究说明,该方法在噪声环境下依然有效,并且不依赖于初始值选择,不需计算梯度,可以在全局范围内搜索。
This paper uses the least square method to make the ultrasonic echo parameter estimation problem converted into the optimization problem,and the improved differential evolution algorithm( DE algorithm) is used to optimize the parameters. Simulation results show that the proposed method is effective in the noise environment,does not depend on the initial value selection,without the need to calculate the gradient,and can be in the global range search.
出处
《微型机与应用》
2016年第4期88-90,共3页
Microcomputer & Its Applications
关键词
DE算法
高斯牛顿法
超声回波信号
信号处理
DE algorithm
Gauss-Newton method
ultrasonic echo signal
signal processing