Abstract
Recursive relationships are developed for an optimal α—β—γ filter having an imperfect predictor. The statistics of the measurement and prediction errors are incorporated in the determination of oprimalα, β and γ and parameters. When the prediction uncertainties and the measurement statistics are white Gaussian, the optimal α—β—γ filter is shown to be equivalent to the Kalman filter.
Indexing Terms:
Additional information
Notes on contributors
K.V. Ramachandra
Ramachandra, (K V) Dr: (b 1939 March 30, Karnataka) Received his MSc (Phy) from Karnataka University and PhD (Maths) from Bangalore University. Presently, he is with Radar Division of Electronics and Radar Development Establishment, High Grounds, Bangalore. He has to his credit around Twenty Five Papers published in leading National and International Journals in various fields of Mathematics, Modelling based on Kalman Filter Theory, Non-parametric detection of Radar Signals and Reliability Theory. His current research interests are in the areas of Radar Signal Processing, Multi-target tracking and Linear Multivariable Control Systems.