Distributed Estimation over Delayed Sensor Network with Scalable Communication

Abstract

This paper proposes a distributed delay-compensated observer for a wireless sensor network with delay. Each node of the sensor network aggregates data from the other nodes and sends the aggregated data to the neighbor nodes. In this communication, each node also compensates communication delays among the neighbor nodes. Therefore, all of the nodes can synchronize their sensor measurements using scalable and local communication in real-time. All of the nodes estimate the state variables of a system simultaneously. The observer in each node is similar to the delaycompensated observer with multi-sensor delays proposed by Watanabe et al. Convergence rates for the proposed observer can be arbitrarily designed regardless of the communication delays. The effectiveness of the proposed method is verified by a numerical simulation.

Distributed Observer over Delayed Sensor Networks for Systems with Unknown Inputs

“…In our past work, we proposed a data-aggregationbased estimation for a sensor network with a communication delay in [10]. In this work, it is assumed that each node is connected by the actuator network without a communication delay.…”

Distributed estimation based on weighted data aggregation over delayed sensor networks

“…For the real-time synchronization over delayed sensor networks, the authors have proposed data aggregation laws with delay compensations [2,1]. In these method, each node compensates communication delays of the received data by the prediction based on the delay compensated observer proposed in [14].…”

“…In these method, each node compensates communication delays of the received data by the prediction based on the delay compensated observer proposed in [14]. In [2], all nodes can estimate the state values by the distributed observers designed by the finite pole assignment technique, where the infinite poles caused by delays are hidden. However, to satisfy the stability condition, in the previously proposed data aggregation laws, the graph topology is limited to tree networks.…”

“…In this paper, to extend the result of [2] to any graph topology, a weighted data aggregation is proposed. The data aggregation over any graph may diverge regardless of the measurement values because old messages continue remaining on a cyclic part of the network.…”

“…The conference paper [2] is a preliminary version of this paper. In this paper, we provide further detailed explanations about the design of weights and practical effectiveness.…”