A recursive linear MMSE filter for dynamic systems with unknown state vector means

In this contribution we extend Kalman-filter theory by introducing a new recursive linear minimum mean squared error (MMSE) filter for dynamic systems with unknown state-vector means. The recursive filter enables the joint MMSE prediction and estimation of the random state vectors and their unknown...

Full description

Bibliographic Details
Main Authors: Khodabandeh, A., Teunissen, Peter
Format: Journal Article
Published: Springer 2014
Subjects:
Online Access:http://hdl.handle.net/20.500.11937/35782
Description
Summary:In this contribution we extend Kalman-filter theory by introducing a new recursive linear minimum mean squared error (MMSE) filter for dynamic systems with unknown state-vector means. The recursive filter enables the joint MMSE prediction and estimation of the random state vectors and their unknown means, respectively. We show how the new filter reduces to the Kalman-filter in case the state-vector means are known and we discuss the fundamentally different roles played by the initialization of the two filters.