Estimate Aggregation with Delay Constraints in Multihop Wireless Sensor Networks

Abstract

Using wireless sensor networks (WSNs) to observe physical processes for control systems has attracted much attention recently. For some real-time control applications, controllers need to accurately estimate the process state within rigid delay constraints. In this paper, we jointly consider state estimation and scheduling problems in multihop WSNs. For accurately estimating a process state as well as satisfying rigid delay constraints, we propose an in-network estimation approach which includes two coupled parts: the estimation operations performed at sensor nodes and an aggregation scheduling algorithm. Our in-network estimation operation performed at intermediate relay nodes not only optimally fuses the estimates obtained from different sensors but also predicts upper stream sensors' estimates which can not be aggregated to the sink before deadlines. Our estimate-based aggregation scheduling algorithm, which is interference-free, is able to aggregate as much estimate information as possible from a network to a sink within delay constraints. We prove the unbiasedness of in-network estimation, and theoretically analyze the optimality of our approach. Our simulation results corroborate the theoretical results.