Title :
An extended Kalman filter for parallel-cascade truncated Volterra systems
Author :
Panicker, Thomas M. ; Mathews, V.John
Author_Institution :
Network Access Div., Rockwell Semicond. Syst., San Diego, CA, USA
Abstract :
This paper presents an extended Kalman filter for truncated Volterra systems that employ parallel-cascade structures. Parallel-cascade realizations implement higher-order Volterra systems as parallel connections of multiplicative combinations of lower-order systems. A computationally efficient variation of the extended Kalman filter obtained by employing a block diagonal approximation of the covariance matrix of the state error vector is also presented in this paper. Experimental comparisons demonstrating relatively small performance degradation for the approximate system over the extended Kalman filter are also included in the paper. Experiments that demonstrate the approximation capabilities of the extended Kalman filter are also included.
Keywords :
Kalman filters; approximation theory; cascade networks; covariance matrices; digital filters; nonlinear filters; approximation capabilities; block diagonal approximation; computationally efficient variation; covariance matrix; extended Kalman filter; higher-order Volterra systems; lower-order systems; multiplicative combinations; parallel-cascade truncated Volterra systems; performance degradation; state error vector; Adaptive filters; Convergence; Covariance matrix; Degradation; Equations; Kernel; Least squares approximation; Matrix decomposition; Nonlinear systems; Resonance light scattering;
Conference_Titel :
Signals, Systems & Computers, 1997. Conference Record of the Thirty-First Asilomar Conference on
Conference_Location :
Pacific Grove, CA, USA
Print_ISBN :
0-8186-8316-3
DOI :
10.1109/ACSSC.1997.680022