论文部分内容阅读
提出了从被加性高斯白噪声污染的混沌时间序列中同时估计混沌系统参数和滤除序列噪声的新方法．并假定产生非线性时间序列的模型已知，但相应的参数未知．这种新方法把对混沌时间序列的参数估计和滤波看作是一种最小化过程，并利用了最速梯度下降方法解决．数值模拟实验表明，新方法要优于现有的方法，是估计混沌系统参数和滤波的一种有效方法．
A new method of simultaneously estimating chaotic system parameters and filtering out sequence noise from chaotic time series contaminated by additive white Gaussian noise is proposed. It is assumed that the model that produces the non-linear time series is known, but the corresponding parameters are unknown. This new method regards the parameter estimation and filtering of chaotic time series as a minimization process and uses the steepest gradient descent method to solve it. Numerical experiments show that the new method is superior to the existing methods and is an effective method to estimate chaotic system parameters and filtering.