Adaptive Pull?CPush Based Event Tracking in Wireless Sensor Actor Networks

Wireless sensor networks have attracted significant interest for various scientific, military, and e-health applications. Recently a new class of sensor networks ??sensor/actor networks?? has been introducing new research challenges due to the unique coordination requirements among sensors and actors. In sensor/actor networks, actors are the nodes that have the capability to move in the field, equipped with powerful devices and can respond to the events of interest. With this capability, autonomous operation of the network is possible without a centralized controlling mechanism. This, however, requires the network to apply cooperative mechanism to decide when and how monitoring is done to track the event and how the event will be responded. In this regard, little work has been done in terms of co-existing Push and Pull data flows in the network. In this paper, we propose an Adaptive Pull?CPush (APP) based Event Tracking approach that allows sensor-to-actor communication as well as actors coordination in response to the events occurred. APP proposes two models of sensors organization: region-based organization (RAPP) and neighbor-based organization (NAPP) to alert nodes in the vicinity of reported event. APP exploits the mobility of actor nodes to form dynamic responsibility clusters, thus ensuring an event specific response to emergencies. Routing in APP is based on Routing by Adaptive Targeting (RAT), which is a delay-constrained geographical routing protocol. Simulation results reveal significant performance improvement in terms of response time and energy conservation.