1: \begin{abstract}

2: 	Classical distributed estimation scenarios typically assume timely and reliable exchanges of information over the sensor network. 

3: 	This paper, in contrast, considers single time-scale distributed estimation via a sensor network subject to transmission time-delays. The proposed discrete-time networked estimator consists of two steps: (i) consensus on (delayed) a-priori estimates, and (ii) measurement update.  The sensors only share their a-priori estimates with their  out-neighbors  over (possibly) time-delayed transmission links. The delays are assumed to be fixed over time, heterogeneous, and known. We assume distributed observability instead of local observability, which significantly reduces the communication/sensing loads on sensors. Using the notions of augmented matrices and Kronecker product, the convergence of the proposed estimator over strongly-connected networks is proved for a specific upper-bound on the time-delay. 

4: 	%Simulations are provided to support our theoretical results.

5: 

6: \keywords  Distributed estimation, consensus, Kronecker product, communication time-delays 	

7: %{\tb Kronecker product: will get back to this later.}

8: \end{abstract}

9: