Abstract:
Presents a method of the parameter estimation of ultrasonic echo based on ant colony algorithm which used for continuous function optimization. According to Gauss echo model, the basic principle and parameter estimation procedures of ant colony algorithm are given. The arrival time and the center frequency are estimated on the conditions with different SNRs (signal to noise ratio) through the numerical simulation. The simulation results show that this method does not depend on the selection of the initial values and can search in a wider scope to obtain the optimal solutions, hence the flaw of the simulation of ultrasonic echo with Gauss-Newton method, which is the Gauss-Newton iteration only can iterative the optimal solutions when the selected initial values are in the vicinity of the true values is avoided. Through the analysis and comparison with Gauss-Newton iteration and simulated annealing, the ant colony algorithm is an effective method to estimate the parameters of ultrasonic echo, and has a higher precision.
