论文部分内容阅读
针对混沌动力学系统时变参数未知的混沌信号,在含有状态噪声的情况下,利用混合卡尔曼滤波提出一种盲估计算法.对未知参数和混沌状态构成的高维状态进行估计,先利用卡尔曼滤波给出线性高斯部分的最优精确估计,剩余部分利用粒子滤波方法给出次优估计,文中详细研究了高斯噪声以及非高斯噪声下的最优重要性函数选取并推导了重要性权重的计算公式,最终基于有效粒子的最小均方误差准则实现了信号的盲估计.仿真结果表明该算法能有效实现含有状态噪声混沌信号的盲估计,并取得了比基本粒子滤波算法更优的性能.
Aiming at the chaotic signal with unknown time-varying parameters in chaotic dynamical systems, a hybrid blind Kalman filtering algorithm is proposed to estimate the chaotic signals with unknown parameters and chaos states. The Kalman Man filter gives the optimal and accurate estimation of the linear Gaussian part, and the rest uses the particle filter method to give the sub-optimal estimation. In this paper, we study the optimal importance function of Gaussian noise and non-Gaussian noise in detail and deduce the importance weight Finally, based on the minimum mean square error criterion of effective particles, a blind estimation of signals is achieved.The simulation results show that this algorithm can effectively achieve the blind estimation of chaotic signals with state noise and achieve better performance than the basic particle filter.