Research On Vehicle Detection Algorithm And Coverage In Wireless Magnetoresistive Sensor Networks

Since the beginning of the new century,China’s economy has taken off,and the number of cars is increasing year by year.At the same time,various traffic problems are emerging in endlessly and the traffic pressure is rising,which leads to the deterioration of traffic conditions.The development of intelligent parking management system can effectively reduce the frequency of traffic problems and induce car owners to park easily.Intelligent parking management system is an important application of vehicle automatic detection technology and the most important research object of static traffic.At present,there are many technical means to realize intelligent parking management system,such as video detection technology,radar detection technology,ultrasonic detection technology,etc.Compared with other detection technologies,the parking management system based on wireless magnetoresistive sensor network technology has the characteristics of low cost,low power consumption,small size,easy installation and maintenance.However,this technology also faces two key problems:（1）wireless magnetic sensor networks will face redundant gateway coverage and high deployment cost when deploying gateways;（2）The parking detection based on magnetic sensor will reduce the detection accuracy in the environment of strong magnetic interference.This thesis mainly studies and solves these two problems:（1）The optimal deployment of gateways in wireless magnetoresistive sensor networks is studied.Through in-depth analysis of the optimal deployment of gateways,a problem model based on set coverage is designed,and a branch cutting method is proposed to solve it.At the same time,two traditional deployment methods are introduced for comparison.Through the analysis of the experimental results of the three methods,the deployment method proposed in this thesis has more advantages.Finally,a simulation deployment system of gateway based on GIS is designed to simulate the above problems and methods,and provide reliable support for the actual deployment of gateway.（2）The problem of anti-interference parking detection in wireless magnetoresistive sensor networks is studied.By analyzing the characteristics of the magnetic interference signals generated by underground high-voltage cables,an anti-interference vehicle detection method based on generalized mathematical morphology is designed.Two structural elements can filter the interference signals well,and the double threshold mechanism is used for detection and determination.In the comparison experiment with two traditional parking detection methods,this method has higher detection accuracy,can well restore the vehicle signal characteristics and accurately describe the parking space status.