Polynomial Estimators for Time Discrete Systems in Hilbert Space,
Abstract
Minimum variance finite polynomial estimators are investigated for time-discrete systems. The estimation problem is formulated in Hilbert space with the optimal estimate as the projection on the linear manifold generated by a finite set of polynomials of the observer. It is shown that this estimate may be partitioned into a predictor and corrector with the corrector being expressed as the product of a weighting vector and measurement residual. The problem of updating the estimator is viewed as two separate problems: incorporating new data and changing the time of the estimate. The first requires the generation of an observation residual and corresponding weighting vector while the second is a prediction or smoothing operation. General expressions for smoothing, filtering and prediction are developed as are formulas for the generation of an orthogonal set set of measurement residuals. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 01, 1972
- Accession Number
- AD0755805
Entities
People
- Wesley K. Masenten
Organizations
- University of California, Los Angeles