摘要
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 for simultaneous filtering and parameter estimation for chao tic time series corrupted by additive white Gaussian noise was presented. In particular, it was assumed that the underlying model of the nonlinear time series was known, but the corresponding parameters were not. The new method treated the problem of parameter estimation and filtering for chaotic time series as a nonlinear minimisation process and solved it by using a steepest descent algorithm. Chaotic time series generated by computer was applied to verify the performance of the new method, which was compared with the existing method. The numerical simulation experiments show that the new method is better than the existing one.
出处
《上海交通大学学报》
EI
CAS
CSCD
北大核心
1999年第1期22-24,共3页
Journal of Shanghai Jiaotong University
基金
国家自然科学基金
