An Improved Spatial-based Fuzzy Logic Event Detecting Algorithm for Wireless Sensor Networks

Baowei Wang, Aoran Zhuang, Huiyu Sun, Tao Li, Xingming Sun
2015 International Journal of u- and e- Service, Science and Technology  
At present, event detection technologies have become an important part in building efficient wireless sensor networks. One of the popular and excellent event detecting algorithms is the improved Neighbor-based Fuzzy Logic algorithm, namely INFLE, which belongs to machine learning technology. However, there may be some false positives when fire has not spread to the detected node. In this paper, we propose an improved spatial-based fuzzy logic event detection algorithm (ISFLE) to dramatically
more » ... to dramatically decrease the probability of false positives before fire has spread to the node. In our proposed ISFLE, we put "Distance" variable to the fuzzy logic system to determine node's final state. The simulation results validate that our proposed ISFLE outperforms traditional INFLE in decreasing the number of false positives. 271 whose radius is 500m. The sensor radio range is set to 100m. Suppose at time 0 fire is ignited from center of the circle, and it spread outwards in all directions at the same speed of 2m/s. Node's sample rate is set to be 1Hz. Figure 5. Experiment Environment Our Fuzzy Logic System The fuzzy logic system for INFLE and ISFLE is displayed through the FuzzyJ Toolkit for Java [9]. Temperature, Smoke membership functions, Distance membership functions [10] and output fire confidence membership functions are showed in Figure6. Furthermore, we define parts of the rules for our fuzzy logic system which are using parts of NFLE and by specialized knowledge which is showed in Table1. One rule consists of five inputs and one output.
doi:10.14257/ijunesst.2015.8.4.24 fatcat:n3iv2k7vgbafnffstb76n7wzdm