1: \begin{abstract}

2:   The unknown inputs in a dynamical system may represent unknown

3:   external drivers, input uncertainty, state uncertainty, or

4:   instrument faults and thus unknown-input reconstruction has several

5:   wide-spread applications.  In this paper we consider delayed

6:   recursive reconstruction of states and unknown inputs for both

7:   square and non-square systems. That is, we develop filters that use

8:   $current$ measurements to estimate $past$ states and reconstruct $past$

9:   inputs. We further derive necessary and sufficient conditions for

10:   convergence of filter estimates and show that these convergence

11:   properties are related to multivariable zeros of the system. With

12:   the help of illustrative examples we highlight the key contributions

13:   of this paper in relation with the existing literature. Finally, we

14:   also show that existing unbiased minimum-variance filters are

15:   special cases of the proposed filters and as a consequence the

16:   convergence results in this paper also apply to existing unbiased

17:   minimum-variance filters.

18: \end{abstract}

19: