摘要
对于未知噪声时变统计特性的时变动态系统，论述了一类基于非平稳噪声参量估计的时变系统参数辨识算法．该算法包括三部分，通过在线状态估计，构造残序列模型，试图从测量信号中分离出非平稳噪声，在线估计噪声时变均值和方差，用于整个改进算法的实现．仿真例子验证了该方法的有效性．
This paper proposed parameter identification algorithm based on nonstationary noise statistics estimation for a class of time-variant dynamic systems with unknown noise statistics. This algorithm consists of three parts: weighted RLS algorithm; on-line state es- timation; residual error model which tyies to separate the nonstationary noise from the measurcment signal and to estimate its time-variant mean-value and variance on-line for realization of the modified algorithm. Simulation study shows that this method is feasible.
出处
《福州大学学报（自然科学版）》
CAS
CSCD
1993年第5期72-76,共5页
Journal of Fuzhou University(Natural Science Edition)
关键词
时变系统
非平稳噪声
参数识别
time-variant systems
nonstationary noise state filter
residual crror model
