Study Of Signal Processing Technology Based On LFMCW Vehicle Flow Monitoring Radar

Recently,with the rapid development of China's economy,the amount of vehicles witnesses a sharp increase in the city,which leads to a higher requirement for the traffic monitoring and management system.In order to alleviate traffic pressure,countries around the world have carried out research on Intelligent Transportation System.And the vehicle detection technology is an important part of the Intelligent Transportation System,mainly to obtain the traffic information of road.Because features,such as accurate and reliable,are needed in the vehicle flow monitoring system,LFMCW Radar,which has a high advantage of precision,reliability and maintenance,can be widely developed and used in vehicle flow monitoring.The paper is concentrated on the research of signal processing method and theory for vehicle flow monitoring radar.The main works of the paper can be summarized as follows:(1)To solve the problem of the frequency band of signal and noise overlapping in LFMCW radar,the wavelet transform in the application of the signal denoising is researched in the paper.And because of the disadvantages of soft and hard thresholding functions in the wavelet threshold method,the paper proposes a new improved thresholding function.The paper uses wavelet threshold denoising method based on hard threshold function,soft thresholding function and the new threshold function proposed to denoise the same real IF signal.The denoising performance of these three functions.The result of experiment shows that the proposed thresholding function can overcome the disadvantages of soft and hard threshold functions and the denoising performance is better.(2)According to the actual needs of LFMCW radar,the adaptive threshold method is researched.The paper proposes a new adaptive detection threshold algorithm based on CFAR.In this algorithm,M reference units are ordered by value firstly.Then m1 the minimum sample spectrum values and the m2 maximum sample spectrum values will be deleted,and the mean value of the rest of sample values is regarded as the estimated value of the power spectrum of clutter background.Finally,the value of threshold is that the estimated value of the power spectrum of clutter background times the threshold coefficient.The power spectrum of signal which has been smoothed is processed by the value of threshold,and the number of vehicle flow can be counted by the result of processing.The result of experiment shows that the precision and stability of the proposed algorithm are preferably.