论文部分内容阅读
传统方法在对超声调制光学信号进行稀疏分解的过程中,存在耗时长、收敛性差等不足,提出改进人工鱼群的超声调制光学信号稀疏分解方法,构建人工鱼群搜索优化模型,模拟鱼群行为搜索最佳原子;针对人工鱼群搜索特点,改进鱼群分布的初值和行为,快速寻找和匹配跟踪过程中每一步信号分解的最优原子,提高方法的收敛速度和稳定性,最终实现超声调制光学信号的稀疏分解。实验证明提出的方法对光学信号分解速度快、全局收敛性好、鲁棒性强。
The traditional method has the disadvantages of long time consuming and poor convergence in the process of sparse decomposition of the ultrasonic modulated optical signal. The method of sparse decomposition of ultrasonic modulated optical signal is proposed to improve the artificial fish school. The artificial fish school search optimization model is constructed to simulate fish school behavior Search for the best atom; search for the characteristics of artificial fish school, improve the initial distribution and behavior of fish, quickly find and match the optimal atom decomposition of each step in the tracking process to improve the convergence rate and stability of the method, and ultimately ultrasonic Sparse decomposition of modulated optical signals. Experimental results show that the proposed method has the advantages of fast decomposition of optical signals, good global convergence and strong robustness.