摘要
本文研究噪声混沌信号的滤波问题。组合信号建模技术,提出了一种基于无先导变换(Unscented Transform,UT)的滤波方法。仿真结果表明,不管混沌系统的参数如何变化,这种算法能都能有效地抑制噪声对混沌系统的影响。与基于扩展卡尔曼滤波器(Extended Kalman filter,EKF)算法相比,在低信噪比的情况下,基于UT的滤波方法都有较好的滤波性能;而在高信噪比的情况下,它有与扩展卡尔曼滤波器基本相同的性能。
This paper addresses the issue of filtering noisy chaotic signals. Combining with the signal modeling technique, a filtering method based on the Unscented Transform (UT) is proposed. Computer simulation indicates that following this approach, the noisy affect on chaotic system can be reduced effectively no mater how the parameters of chaotic system vary. In comparison with the EKF-based method, UT-based method has better filtering performance under the case of low signal to noise ratio (SNR), and has similar performance under the case of high SNR.
出处
《电路与系统学报》
CSCD
2004年第4期58-62,共5页
Journal of Circuits and Systems
基金
西南师范大学博士基金资助项目(220-413604)
教育部回国留学人员专项基金资助项目(130501)
关键词
混沌信号
滤波
无先导变换
扩展卡尔曼滤波器
自回归模型
chaotic signal
filtering, unscented transform
extended Kalman filter
autoregressive mode