论文部分内容阅读
混沌信号所固有的非周期、宽带频谱和对初值极度敏感等特性使得对这类信号进行盲分离极为困难.针对这一问题,提出一种新的盲分离方法,该方法通过相空间重构来构造代价函数,将混沌信号的盲分离转化为一个无约束优化问题,并利用人工蜂群算法进行求解.不同于现有的独立成分分析方法仅使用混合信号的统计特性来解决分离问题,该方法能充分利用混合信号内在的动态特性,因而在处理混沌信号这种确定性信号时能获得更好的分离效果.此外,正交矩阵的参数化表示有效地降低了盲分离问题的复杂性,使优化过程能快速收敛.实验结果表明,该方法具有较快的收敛速度和较高的数值精度,在分离混沌信号时其整体性能优于现有的几种盲分离方法.同时,在分离混沌-高斯混合信号的实验中该方法也展现出优异良好的性能,这表明该文的方法有应用潜力.
The characteristics of chaotic signal such as aperiodic, broadband spectrum and extreme sensitivity to initial value make it difficult to blindly classify these signals.A new blind separation method is proposed to solve this problem, which is based on phase space reconstruction To construct the cost function, the blind separation of chaotic signals is transformed into an unconstrained optimization problem, and the artificial bee colony algorithm is used to solve the problem.Compared with the existing independent component analysis methods, the statistical properties of mixed signals are used to solve the separation problem, The method can make full use of the intrinsic dynamic characteristics of the mixed signal and thus obtain better separation effect when dealing with the deterministic signals such as chaotic signals.In addition, the parametric representation of the orthogonal matrix effectively reduces the complexity of the blind separation problem, So that the optimization process can converge rapidly.The experimental results show that the proposed method has faster convergence speed and higher numerical precision and its overall performance is better than the existing methods in separating chaotic signals.At the same time, - Gaussian mixed signal experiments show that the method also showed excellent good performance, indicating that the method of the paper has potential applications.