Networked Estimation with an Area-Triggered Transmission Method

This paper is concerned with the networked estimation problem in which sensordata are transmitted over the network. In the event-driven sampling scheme known aslevel-crossing or send-on-delta, sensor data are transmitted to the estimator node if thedifference between the current sensor value and the last transmitted one is greater than agiven threshold. The event-driven sampling generally requires less transmission than thetime-driven one. However, the transmission rate of the send-on-delta method becomeslarge when the sensor noise is large since sensor data variation becomes large due to thesensor noise. Motivated by this issue, we propose another event-driven sampling methodcalled area-triggered in which sensor data are sent only when the integral of differencesbetween the current sensor value and the last transmitted one is greater than a giventhreshold. Through theoretical analysis and simulation results, we show that in the certaincases the proposed method not only reduces data transmission rate but also improvesestimation performance in comparison with the conventional event-driven method.