A Method of Target Identification with UWB Based on S-Transform and Improved Artificial Bee Colony Algorithm

摘要

Ultra-wideband signal (UWB) has been used in communication, location, and identification. In this paper, a novel target identification method is proposed. The UWB received signals is processed by time-frequency analysis method: S-transform. S-transform is an extension of short time Fourier transform and continuous wavelet transform, and it has good time-frequency characteristics. Then we use improved artificial bee colony (ABC) algorithm to optimize the penalty factor and kernel parameter of support vector machine (SVM), and finish the target identification. In view of the basic artificial bee colony algorithm has the problem of slow convergence speed. We propose a probability selection method based on quadratic function to optimize the algorithm.