RENDEZVOUS: towards fast event detecting in wireless sensor and actor networks

Fast event detection is important in wireless sensor and actor networks (WSANs) since actors can perform appropriate actions which sensor nodes are not capable to do. While WSANs inherits the typical constrains of WSNs such as energy and computation limitations of sensor nodes. In this paper, we propose a fast event detecting algorithm named RENDEZVOUS to accelerate the actor’s event detecting process while keep the energy consumption of sensor nodes as minimum. When design RENDEZVOUS, we first study the mobility control of a actor to help the actor move around close to a event by using Reinforcement Learning techniques with collected sensory data. We then design a scheme to search nearby actors from the event side inspired by a searching behavior of desert ants. By both perform search actions from sensor side and actor side, the proposed algorithm can achieve fast event detecting with neglect-able additional energy cost on sensors side. Extensive simulation results demonstrate the efficiency of RENDEZVOUS.