Simultaneous input and state estimation for linear discrete-time stochastic systems with direct feedthrough

In this paper, we present an optimal filter for linear discrete-time stochastic systems with direct feedthrough that simultaneously estimates the states and unknown inputs in an unbiased minimum-variance sense. We argue that the information about the unknown input can be obtained from the current time step as well as the previous one, making it possible to estimate the unknown input in different ways. We then propose one variation of the filter that uses the updated state estimate to compute the best linear unbiased estimate (BLUE) of the unknown input. The comparison of the new filter and the filters in existing literature is discussed in detail and tested in simulation examples.