【摘要】 信号的稀疏表示在信号处理的许多方面都有重要的应用,但稀疏分解计算量十分巨大,难以被推广而实现产业化。混沌是一种普遍的非线性现象,具有随机性、遍历性和内在规律性的特点,混沌运动能在一定范围内按其自身的规律不重复地遍历所有状态。因此,如果利用混沌变量进行优化搜索,无疑会比随机搜索更具优越性。本文利用变尺度混沌优化方法在优化搜索过程中不断缩小搜索空间,快速寻找匹配追踪(MP)过程中每一步的近似最佳原子,提高信号稀疏分解的速度,算法的有效性为实验结果所证实。更多还原

【Abstract】 Sparse representation of signals has many important applications in signal processing,but the computational burden in the signal sparse decomposition process is so huge that it is almost impossible to apply it to industrialization.Chaos is a universal nonlinear phenomenon with the stochastic property,ergodic property and regular property,and the movement of chaos has the ability to ergodic all the statements in the area with its own regularity without repetition.So if we put the chaos variables into practice to achieve chaos optimal search,it is no doubt that this algorithm will be much more superior than random search.In this paper,the mutative scale chaos optimization algorithm is implemented to fast search for the approximately optimal atom at each step of Matching Pursuit(MP),and the speed of signal sparse decomposition is improved a lot.The validity of this algorithm is proved by experimental results.更多还原