利用粒子群算法实现PPS信号的稀疏分解  

Sparse Decomposition of Polynomial Phase Signals with Particle Swarm Optimization

作　　者：李越雷[1] 张天骐[1] 黄铫[1] 蒋世文[1] 

机构地区：[1]重庆邮电大学信号处理与片上系统研究所,四川重庆400065

出　　处：《计算机仿真》2010年第2期200-203,共4页Computer Simulation

摘　　要：针对在分析高阶多项式相位信号(PPS)时,Wigner-Ville分布(WVD)的交叉项使得时频分布图变得难以解释,为了提高信号计算速度和数据提取精度,采用基于匹配追踪(MP)算法的信号稀疏分解来抑制交叉项,但是稀疏分解计算量大,难以应用在实时信号处理。将粒子群优化算法用于稀疏分解的最优匹配原子的搜索,能降低稀疏分解复杂度,同时减少稀疏分解的超完备字典对存储空间的占用,可以提高用稀疏分解理论进行信号处理的计算效率,满足或接近实时性的要求。计算机仿真结果证实了方法的有效性。For the analysis of high order polynomial phase signals, the figure of time - frequency distribution becomes difficult to understand due to cross - terms in the Wigner - Ville distribution （WVD）. Cross - term suppres- sion is obtained by using spare decomposition based on matching pursuit（ MP）, but the computational burden in signal sparse decomposition process is so huge that it is almost impossible to apply it to real time signal processing. To reduce complexity of sparse decomposition and space of memory, particle swarm optimization is used in searching the best atom. Particle swarm optimization can increase the efficiency processing signal using sparse decomposition, and then this method can meet （ or near） the request of real time. The validity of this method is proved by experimental results.

关 键 词：稀疏分解 匹配追踪 粒子群优化算法 多项式相位信号 

分 类 号：TN911.7[电子电信—通信与信息系统]