An Efficient Algorithm for Preserving Events' Temporal Relationships in Wireless Sensor Actor Networks

Abstract

This paper proposes an event ordering algorithm for wireless sensor actor networks (WSANs) that could be applied for monitoring critical conditions (such as fires, explosions, toxic gas leaks etc.) in order to ensure the correct interpretation of events. We propose modifications to the ordering by confirmation event ordering protocol for WSANs by introducing clustering into the network's topology. The objectives of the proposed modifications are a reduced number of messages, energy efficiency, scalability and reduced latency. At the same time, we propose a hybrid synchronization scheme for the clustered topology, in which local time scales are used at the level of clusterheads. The clusterheads are synchronized with each other by the actor node using the reference broadcast synchronization technique, while the nodes inside clusters are synchronized with the round trip synchronization technique. The synchronization scheme aims at preserving energy and reducing network delay and it is better suited for the resource sparseness of wireless sensor networks as opposed to methods that use global time scales. Moreover, our proposed algorithm uses the message exchange necessary for event ordering and routing protocols for time synchronization purposes by piggybacking synchronization pulses and replies on these messages, thus reducing the additional traffic needed for time synchronization. In this paper, we present our protocol, discuss its implementation and provide its proof of correctness.