High-speed Railway Track Settlement Measurement Based On RCS Statistical Characteristic And Adaptive Frequency Estimation Algorithm

The railway track settlement affects the stability and comfort of passengers and even threatens the safe operation of trains.Therefore,the research on track settlement measurement has attracted much attention.With the continuous development of radio frequency sensing technology,millimeter-wave radar is widely used in many fields due to its millimeter-level distance measurement accuracy.In response to the requirements of track settlement measurement for high-speed railways,this paper realizes track recognition based on the Radar Cross Section(RCS)characteristics of each target collected by millimeter-wave radar.On this basis,a high-resolution frequency estimation algorithm is used to realize the railway track settlement measurement.The main contents and innovations of the paper are as follows:first of all,based on the frequency modulated continuous wave(FMCW)millimeter-wave radar,a method for recognizing the railway track based on RCS characteristics is proposed.Collected RCS data sets of multiple objects in the track settlement environment,analyzed the effective features of RCS,and used wavelet transform to extract RCS characteristics to facilitate classifier training.Thus,through the classification and recognition algorithm,the track recognition in the dynamic and static mixed scene realize.Then,the range principle and accuracy of FMCW radar are analyzed,and the problem of high-precision ranging is transformed into the problem of high-precision frequency estimation for the frequency of intermediate frequency signals.Besides,the improved ACZT,RootMUSIC,and ZoomFFT algorithms are proposed to achieve high-resolution frequency estimation.The complexity of the above three algorithms is analyzed,which reduces the computational complexity of the algorithm while realizing high-resolution frequency estimation.An experimental platform for railway track settlement measurement is built,and a large amount of experimental data was obtained by conducting experiments under different conditions.The factors affecting track recognition and settlement measurement are explored.Moreover,the performance of the track recognition algorithm and the track settlement measurement method applied in the paper are compared.The experimental results show that the accuracy of track classification and identification is at least 95%,and the accuracy of track settlement measurement exceeds 0.5mm,which completely meets the accuracy requirements of the railway system.