Experience-based learning for identifying sub-regions in Wireless Sensor Networks

Abstract

In this paper, we propose a novel mechanism to calculate sub-regions (overlapped areas) in Wireless Sensor networks (WSNs). As the major part of WSN tasks are monitoring and reporting events in sensors' sensing range, those reported events' locations can be used, by means of convex hulls, to accumulatively learn the boundaries of those overlapped areas between the ranges of sensors. Although the proposed mechanism targeted clustered networks, the mechanism also can be used with non-clustered networks. Besides, the proposed method provides two levels of abstraction, the first level is the selection of the proper algorithm to calculate convex hulls, and the second is the selection of the clustering algorithm at implementation time. The main contribution of this work is to provide a new perspective to solve this problem in WSNs and new avenues for future research.