Joint estimation of fast-updating state and intermittent-updating state

Abstract

This paper formulates a joint estimation problem of fast-updating state and intermittent-updating state in multi-rate systems. The original multi-rate system is first transformed into a single-rate one. Since the direct use of Kalman filtering method on the lifted system will result in the Kalman smoother, the causality constraints must be taken into account in the filter design. Then, based on the lifted system a multi-rate filter (MRF) with causality constraints is derived in the linear minimum mean squared error (LMMSE) sense using the orthogonality principle. A numerical example is given to show the effectiveness of the proposed filter.