A Differential Evolution Based Time-Frequency Atom Decomposition for Analyzing Emitter signals

作者:Zhang Gexiang*; Cheng Jixiang
来源:6th International Conference on Modeling Decisions for Artificial Intelligence, 2009-11-30 to 2009-12-02.

摘要

This paper discusses the use of time-frequency atom decomposition based on a differential evolution to analyze radar emitter signals. Decomposing a signal into an appropriate time-frequency atoms is a well-known NP-hard problem. This paper applies a differential evolution to replace the traditional approach, a greedy strategy, to approximately solve this problem within a tolerable time. A large number of experiments conducted on various radar emitter signals verify the feasibilities that the time-frequency characteristics are shown by using a small number of decomposed time-frequency atoms, instead of traditional time-frequency distributions.