A Linear Filter for Discrete Systems with Correlated Measurement Noise
This paper introduces an optimal linear filter for discrete systems with correlated measurement noise by generalized least square method which is novel in its structure, its derivation and its simplicity. The equations reduce to the standard Kaiman filter equations when the measurement noise is independent. The new filter avoids the increased order and other complexities of previously proposed methods particularly those based on augmented state and differencing approaches.
KeywordsKalman Filter Riccati Equation Discrete System Correlate Noise Linear Filter
Unable to display preview. Download preview PDF.
- 2.Bryson Jr., A. E. and Johansen, D. E., “Linear Filtering for Time-Varying Systems Using Measurements Containing Colored Noise,” IEEE Trans. on Automatic Control, pp. 4–10, January 1965.Google Scholar
- 6.Kailath, T. and Geesey, R., “Covariance Factorization,” Proceedings of 2nd Asilomar Systems Conference, 1968.Google Scholar
- 7.Johnson, D. J., “A General Linear Sequential Filter,” Paper Presented at 1970 JACC, Atlanta, Georgia, June 24–26, 1970.Google Scholar
- 8.Bucy, R. S., Rappaport, D. and Silverman, L. M., “Correlated Noise Filtering and Invariant Directions for the Riccati Equation,” To be published in IEEE Trans. on Automatic Control.Google Scholar
- 9.Bucy, R. S., Rappaport, D. and Silverman, L. M., “A New Approach to Singular Discrete Time Filtering and Control,” Proceedings of the 3rd Hawaii Conf. on Systems Science, Honolulu, Hawaii, February 1970.Google Scholar
- 11.Bucy, R. S. and Joseph, P. D., Filtering for Stochastic Processes with Applications to Guidance, Interscience Publishers, New York, 1968.Google Scholar
- 12.Goldberger, A. S., Econometric Theory, John Wiley and Sons, Inc., New York, 1964, pp. 232–233.Google Scholar