State estimation for a class of nonlinear systems with multi-package transmission

In this paper, the state estimation problem for a class of nonlinear systems with multiple channels and correlated noises is studied within the same framework of Extended Kalman Filter (EKF). In networked systems, when sensors are distributed in a large spatial area and multiple channels are employed to transfer data from different sensors, parts of the measurements may be lost at different rates. In the present work, a diagonal matrix is utilized to describe this phenomenon and an unbiased sub-optimal filter is constructed recursively in least mean square sense. An illustrative example is provided and the comparison of the results between our method and the classical EKF shows the effectiveness of the proposed approach.