Selection of the optimized window for the generalized s-transform using the AIW-PSO algorithm

Abstract

This work proposes a method for the selection of the optimized window of analysis for the Generalized S- Transform (GST), through an adaptation in the Particle Swarm Optimization (PSO) algorithm. The new version of the algorithm, called Adaptive Individual Inertia Weight Based on Best, Worst and Individual (AIW-PSO), uses an inertia weight, where the performance of the best and the worst particle are employed in the adaptive process. The GST is applied to the signal by using AIW-PSO, whose objective function to be minimized is the concentration of energy that will provide the proposed window selection. To evaluate the efficacy of the proposed method, two synthetic signals were analysed by the GST, based on parameters r, m, p and k, and compared with the standard version and an optimized version of the S-Transform (ST) available in the literature. It was found that the GST, optimized via the AIW-PSO algorithm, provided a better response in the Time-Frequency Representation (TFR) when compared to the standard non-ideal ST window, as well as an additional gain in energy concentration relative to the optimized version approach.