NONPARAMETRIC APPROACH TO IDENTIFYING NARX SYSTEMS

• 作 者: Qijiang SONG Han-Fu CH1N
• 机 构: Key Laboratory of Systems and Control, Institute of Systems Science, Academy of Mathematics and Syst
• 出 处: 系统科学与复杂性学报（英文版）
• 发表日期: 2010年1期
• 关键词: α-mixing geometrically ergodic Markov chains NARX nonparametri1 recursive esti-m

This paper considers identification of the nonlinear autoregression with exogenous inputs (NARX system). The growth rate of the nonlinear function is required be not faster than linear with slope less than one.The value of f(.) at any fixed point is recursively estimated by the stochastic approximation (SA) algorithm with the help of kel functions. Strong consistency of the estimates is established under reasonable conditions, which, in particular, imply stability of the system. The numerical simulation is consistent with the theoretical analysis.