超声回波参数的高斯与牛顿迭代法计算

The calculation of ultrasonic echo parameter by Gauss-Newton iterative method

应用超声回波模型,对检测目标的超声回波进行模拟,在模拟过程中应用高斯与牛顿法迭代出估计参数.结果表明,该算法能以较少的迭代次数计算出向量参数,但对模型中到达时间初值设定较为敏感,这可用常规方法(互相关或小波变换、遗传算法等)解决.

The ultrasonic echo model is applied to simulate the ultrasonic echo of detection object and the Gauss-Newton iterative method is applied in the simulation process to estimate the parameters. The result shows that the algorithm can calculate the .vector parameter with few iterative times. The convergence of the algorithm is sensitive to the initial value of arrival time can be solved by some conventional methods, such as cross-correlation, wavelet transform and genetic algorithms.