Strategies for Sensor Data Aggregation in Support of Emergency Response
Xianping Wang, Aaron C. Walden, Michele C. Weigle, Stephan Olariu · 2014
Sensor networks deployed in support of emergency response applications must provide timely and accurate reports of detected events. Aggregation of sensor data is required to accomplish this in an efficient manner. The aggregation problem is complicated by the fact that the perceived value of the data collected by the sensors deteriorates, often dramatically, over time. Individual sensors must determine whether to report a perceived event immediately or to defer reporting until the confidence has increased after aggregating data with neighboring nodes. However, aggregation takes time and the longer the sensors wait, the lower the value of the aggregated information. Our main contribution is a formal look at various novel aggregation strategies that relate aggregation decisions to the ensuing value of information. Our results suggest natural thresholding strategies for aggregating the information collected by groups of sensors. Our results are applicable to emergency response and other similar homeland security and military application domains where there is a strong need to model not only timely aggregation of data collected by individual actors, but also the dynamics of this aggregation. Extensive simulations have confirmed the accuracy of our theoretical predictions.