Improved Kalman Filtering for Systems with Randomly Delayed and Lost Measurements

In this paper, a novel Kalman filter is developed for discrete-time linear systems with intermittent observations. The measurement data latency and dropout in transmission from the sensor to the filter is modeled by a group of Bernoulli distributed random variables. The minimum variance filter is designed when the measurement data packets are/are not time stamped by the reorganized innovation approach. Moreover, the steady-state behavior of the proposed Kalman filter is investigated. Finally, simulation results are presented to illustrate that the suggested estimator leads to the remarkably improved performance compared with the previously developed approaches.