We propose D-FLER, a distributed, general-purpose reasoning engine for WSN. D-FLER uses fuzzy logic for fusing individual and neighborhood observations, in order to produce a more accurate and reliable result. Thorough simulation, we evaluate D-FLER in a fire-detection scenario, using both fire and non-fire input data. D-FLER achieves better detection times, while reducing the false alarm rate. In addition, we implement D-FLER on real sensor nodes and analyze the memory overhead, the numerical accuracy and the execution time.
展开▼
