Design And Implement Of Event Detecting Algorithm Based On Fuzzy Logic In Wireless Sensor Networks

Wireless sensor networks are composed of a number of small, inexpensive devices, called sensor nodes, which are equipped with sensing, processing, and communication capabilities. These nodes can communicate with each other wirelessly to form a multi-hop self-organizing network.At present, event detection technologies have become an important part in building efficient wireless sensor networks. A way to identify whether the detecting scheme is good or not is that the scheme can effectively reduce the false negatives and false positives for the event. False negatives mean that the scheme cannot detect the event when event has happened and false positives mean that the scheme detect the event by mistake when event has not happened. We propose the detecting schemes to detect fire in buildings which mainly focus on reducing the false negatives and false positives. The scheme can detect fire based on the fire confidence for the node which represents the possibility of fire on the node. However, traditional neighbor-based fuzzy logic event (NFLE) detecting scheme may produce some false negatives in high-precision fire detecting networks when fire has not spread to most of its neighbors. So we come out with an improved neighbor-based fuzzy logic event detecting scheme. Besides, we further decrease the false positives rate by proposing an improved spatial-based fuzzy logic event detecting scheme when the selected nodes are farther from the detected node.This thesis to complete the following work:(1) To solve the problem of false negatives in traditional neighbor-based fuzzy logic event detecting scheme, we propose an improved neighbor-based fuzzy logic event (INFLE) detecting scheme.(2) To solve the problem of false positives in INFLE scheme when the selected nodes are farther from the detected node, we propose a spatial-based fuzzy logic event (ISFLE) detecting scheme.