ON MINIMUM-MEAN-SQUARED ERROR NONLINEAR FILTERING.
Abstract
The study is concerned with the minimum-mean-squared error (MMSE) nonlinear filtering of signals in additive noise. Methods of analysis, representation, and classification of non-linear systems are treated. Integral representations and their applicability to random inputs and MMSE filtering are emphasized. The general MMSE nonlinear filtering problem is discussed briefly; the restriction of the filter to general classes of systems or input processes is considered. Usually the general solutions to the problem are difficult and in many cases are impractical; therefore restriction to special classes of systems and special classes of input processes is investigated. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Aug 01, 1966
- Accession Number
- AD0645744
Entities
People
- A. H. Haddad
- J. B. Thomas
Organizations
- Princeton University