摘要
研究了基于Volterra模型的非线性系统的鲁棒自适应辨识问题 .针对Volterra系统辨识时输入输出观测数据均受噪声污染的情况 ,建立了基于Volterra模型的鲁棒Volterra总体均方最小自适应辨识算法 .该算法应用梯度下降原理 ,通过对梯度的修正 ,有效地提高了算法的鲁棒性 .仿真结果表明 ,在低信噪比 ,或使用较大学习因子的情况下 ,该算法的收敛性能明显优于其他算法 ,便于实际应用 .
Adaptive identification for nonlinear Volterra system is researched. In allusion to the Volterra system identification in which the input and output signals are all corrupted by noise, a robust Volterra total least mean square adaptive identification algorithm is presented based on Volterra series model. This algorithm is established according to the steepest descent principle, and its robust performance is effectively improved by modifying gradient. The simulation results show that the convergence performance of the presented algorithm is better than those of other algorithms when signal-noise-ratio (SNR) is lower, or a larger learning factor is used. This new algorithm could be used in actual application.
出处
《西安交通大学学报》
EI
CAS
CSCD
北大核心
2001年第10期1024-1028,共5页
Journal of Xi'an Jiaotong University
基金
陕西省重点科技项目 (2 0 0 0K0 8-G6 )
