Adaptive Identification of Nonlinear Systems Operating in Closed-loop

本研究的目的是提出有效的在线自适应方法的参数识别的时变线性和非线性元件在闭环系统中运行。（1）在控制扬声器到参考传声器之间存在不确定耦合路径的情况下，首次提出了一种理想的自适应有源噪声控制方案，在不需要识别从控制扬声器到误差传声器的次级路径动态特性的情况下，抑制误差传声器的初级噪声。该方法的有效性也已在使用空气管道的主动降噪中得到验证。(2)针对双质量电机速度反馈控制系统中的线性和非线性摩擦和阻尼因子辨识问题，提出了一种稳定的自适应控制算法。(3)针对OFDM系统中大功率放大器的线性和非线性失真，提出了自适应预失真方案。(4)使用阵列传感器的到达方向估计方法已经在多径情况下被开发，并且也被应用于消除来自发射机天线的耦合信号。

The aim of this research is to present efficient on-line adaptive methods for identification of parameters of time varying linear and nonlinear elements operating in closed-loop systems. The obtained results are summarized as follows : (1) A perfect adaptive acitve noise control scheme has been given first to suppress primary noise at an error microphone without identification of any secondary path dynamics from artificial control loudspeaker to the error, in a situation when there is also an uncertain coupling path from the control loudspeaker to a reference microphone. Its effectiveness has also been validated in active noise attenuation using an air duct. (2) The stable adaptive control algorithm was derived to identify linear and nonlinear friction and damping factors in two-mass motor velocity feedback control systems. (3) Adaptive predistortion schemes have been given to compensate for linear and nonlinear distortions of high power amplifier used in OFDM systems. (4) Direction-of-arrival estimation methods using an array sensor have been developed in of multipath situations, and also applied in cancellation of coupling signals from a transmitter antenna.