A network reliability oriented event detection scheme for Wireless Sensors and Actors Networks

Wireless Sensors and Actors Networks (WSANs) emphasize the necessity of accurate event detection in order to achieve ubiquitous monitoring in realistic environments and perform appropriate actions on them. One of the most significant applications is health monitoring. This paper is aiming to propose a fuzzy logic system based on multi-sensor vital and environmental data, in order to detect distress situations. Also, a data accuracy fuzzy system is presented in order to estimate the network confidence. Both systems have been evaluated in order to demonstrate their validity and usefulness. The outputs of the aforementioned systems are associated so as to trigger the optimal acting tasks. The proposed data fusion scheme aims at improving the reliability of the system and can be adaptively modified and further enhanced in terms of more sensor readings, number of rules etc.

I. Introduction

In recent years, Wireless Sensors and Actors Networks (WSANs) have been used in a wide variety of applications including home automation, intelligent industrial control, environmental monitoring and healthcare applications. WSANs consist of stations with different resources and capabilities. The sensing and acting tasks should be examined cooperatively. Thus coordinating mechanisms are necessary for the sensor-actor and the actor-actor negotiation. A very challenging task in this research area, that has attracted a lot of interest, is identifying the communication protocols and the network architecture. In addition, distributed event detection is one of the critical issues in WSANs [1]. A decision model is essential, so as to allow the correct classification of an event and increase the robustness of the system especially in demanding environments such as industrial networks. The classification of an event can be defined as the process of observing and evaluating an event using multiple sensor nodes. The final decision is made by fusing local values to increase event detection accuracy and reduce communication overhead [2].