Abstract
This paper presents output voltage estimation of metal-oxide- semiconductor field effect transistor (MOSFET) circuit using extended Kalman filter (EKF), in which, the nonlinear system dynamics has been modeled using Kronecker product. Input voltage is modeled as an Ornstein- Uhlenbeck (O.U.) process to account both, the white noise and Brownian process. State space model of the circuit has been obtained using Enz-Krummenacher-Vittoz (EKV) model of the MOSFET circuit and Kirchhoff’s current law (KCL). The proposed method has the following advantages (i) It can be used for any mode of transistor operation besides near the quiescent point region. The nonlinearity in saturation can be considered in the proposed method. (ii) The method can be used for large amplitude input signal. For small amplitude input signal, Kalman filter (KF) can be used, which results in inaccurate estimation due to linearization of nonlinear system. (iii) The method presents real time parameter estimation, as EKF has been used for nonlinear system. It is able to track the parameters, when they are slowly changing with time. (iv) Use of Kronecker product presents more accurate representation of nonlinear system. The proposed method has been compared with Kronecker based wavelet transform (WT) representation of the system. The wavelet coefficients have been obtained using perturbation technique. Finally, the least mean square (LMS) has been applied to the wavelet transform representation of the system for parameter estimation. Simulation results validate the performance of the proposed method.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call