A High-Performance Vehicle Detection Algorithm for Wireless Sensor Parking Systems

This paper introduces a high-performance vehicle detection algorithm for a parking management system based on wireless sensor technology. The parking lot is represented as a wireless sensor network cluster modeled as an m × n grid in a rectangle of size I × h. A wireless sensor node is deployed in each grid and one gateway is set for the cluster. The status of each parking space is detected by a sensor node and the result is reported to the gateway. To reduce the false positive rate caused by various sources of interference, we developed a high-performance vehicle detection algorithm to filter out the noises from the passing-by cars and the environment. The performance of the algorithm is verified through extensive simulation. The three scenarios of low, medium and high interference levels are simulated. Results suggest that our proposed vehicle detection algorithm can effectively detect the status of parking lot with zero false positive rate at the low and medium interference levels, and 3.3% at high interference level.