(1.裝備學院研究生院，北京 101416；2.裝備學院光電裝備系，北京 101416）

　　摘 要：在目前的通信領域中，跳頻信號的應用越來越廣泛，而信號的稀疏表示在信號處理的許多方面都有著重要的應用。將跳頻信號稀疏分解后得到的稀疏表示可以方便的應用到跳頻信號的參數估計及后續處理中去。本文針對目前基于MP（Matching pursuit）的跳頻信號稀疏分解中運算量巨大的問題，采用遺傳算法（GA）快速地尋找MP算法中每一步分解的最佳原子，提高了信號稀疏分解的速度。仿真分析驗證了此方法的有效性和適用性。
　　關鍵詞：跳頻信號 遺傳算法 稀疏表示 稀疏分解 Matching pursuit（MP）
Using genetic algorithm to realize frequency hopping signal sparse decomposition based on MP

Yi QIN1, Weigang ZHU2, Qiquan SHU1
1. Postgraduate School, Academy of Equipment, Beijing, China
2. Department of Optical and Electrial Equipment, Academy of Equipment, Beijing, China
　　
 Abstract：In the current communication filed, the application of frequency hopping signal become more and more widely and sparse representation of signals has found important applica- tions in signal processing. Through sparse decomposition ,the sparse representation of frequency hopping signal can be easily applied to the hopping signal parameter estimation and subsequent processing. In this paper, to the question of the huge amount of computation in frequency hopping signals sparse decomposition based on MP(Matching pursuit) , genetic algorithms (GA) is used to effectively search in the dictionary of atoms for the best atom at each step of MP, and the speed of signal sparse decomposition is improved a lot. Simulation and analysis verify the validity and applicability of this method . 
 Key Words：frequency hopping signal  genetic algorithms  sparse representation  sparse decomposition  Matching pursuit(MP)

　　
參考文獻
[]Mallat S, Zhang Z. Matching pursuit with time-frequency dictionaries[J].IEEE Trans. on Signal Processing,1993,41(12):3397-3415.
[] 張文耀.基于匹配跟蹤的低位率語音編碼研究[D].北京:中國科學院研究生院軟件研究所，2002.
[]Arthur P L, Philipos C L. Voiced/Unvoiced Speech Discrimination in Noise Using Gabor Atomic Decomposition[C].Proc. of  IEEE 1CASSP. Hong Kong,2003; I, 820-828.
[] 梅文華，王淑波，邱永紅.跳頻通信[M].國防工業出版社,2005.
[] Davis L. Handbook of Genetic Algorithms[M].New York: Van Nostrand Reinhold. 1991.