Research On Anti-interference Parking Detection Algorithm Based On Magnetic Sensor

In recent years,with the rapid development of economy,the number of motor vehicles in China has increased year by year,and the traffic pressure has increased dramatically.Especially during peak hours,the tightening supplies of parking space in the downtown has led to increasingly terrible traffic parking problems.The intelligent parking management system provides a solution to the problem of difficult parking.Real-time parking detection is an important part of the intelligent parking management system.The wireless magnetic sensor network detects ferromagnetic substances by magneto-resistance effect to realize vehicle detection.This technology has the advantages of high detection accuracy,low cost,flexible deployment,and convenient maintenance,and it has been gradually applied to intelligent parking management systems.There are a lot of research results about the parking detection algorithm of wireless magnetic sensor network at academic and industry circles.However,the technology faces multi-source interference problems,as follow:1)Adjacent interference: Adjacent interference comes from vehicles in adjacent parking space.Some vehicles have a lot of ferromagnetic material,which can generate wide range of magnetic field influence.This effect will be misinterpreted as a vehicle signal by the adjacent space sensor.2)Subway interference: Subway magnetic interference mainly comes from the power supply system of subway,a part of current which come from overhead conductors leaks to the ground with the rails.This stray current will cause magnetic interference after contact with the ground.3)Parking lock interference: The parking lock lever is a ferromagnetic substance,which will generate a magneto-resistance effect during working,and the magnetic interference will be captured by the parking lock of the built-in sensor.Existing vehicle detection algorithms do not take into account the phenomenon of magnetic interference.The detection accuracy of traditional vehicle detection algorithms decreases by the influence of multi-source interference.The main research of this paper is anti-interference parking detection algorithms based on magnetic sensors.For the above three types of interference,we have proposed three anti-interference parking detection algorithms:1)In the light of the adjacent interference,this paper proposes a data fusion algorithm based on waveform similarity.This algorithm reduces the calculation amount of waveform comparison,and filters near interference through multi-sensor collaborative decision-making.2)In the light of the subway magnetic interference,this paper proposes a filtering algorithm based on mathematical morphology,this algorithm filters the magnetic interference caused by stray current and corrects the magnetic baseline.3)In the light of the parking lock interference,this paper proposes a filtering algorithm based on variance sequence to extract vehicle signals,filter parking lock interference and describe the parking state through the state machine.Experimental results show that the algorithm in this paper has a good suppression effect on magnetic interference.Compared with the traditional algorithm,the algorithm in this paper improves the accuracy of the vehicle detection and accurately describes the status of the parking space.