Adaptive Time-Frequency Parameterization of Frequency-Hopping Signals Based on Evolutionary Algorithm

Matching pursuit algorithm extracting the time-frequency characteristics of signal has been applied in many fields. High computer complexity is a bottle-neck, especially in the high dimensions of the search space. In this paper, genetic algorithm and particle swarm optimization is used to solve this problem. Two decomposition methods named particle swarm optimization matching pursuit (PSO-MP) and genetic algorithm matching pursuit (GA-MP) are proposed for time-frequency analysis of frequency hopping signals. Experiment results proved the validity and feasibility of the approaches. Compared to GA-MP algorithm, PSO-MP algorithm could choose more precise atom parameters and has higher convergent speed as to the average process time.