This paper is devoted to the design of a decentralized optimal batch LQ state observer for state estimation of large-scale interconnected systems, well suited for implementation on a sensor network. The here-proposed approach relies on both the use of an augmented Lagrangian formulation and a price-decomposition coordination algorithm. The state estimation of an open-channel hydraulic system illustrates the effectiveness of this approach and is used to provide a comparison with alternative methods. (c) 2014 Elsevier Ltd. All rights reserved.