Back to Search Start Over

Estimating Individual Vehicle Speed by FMCW Radar Detector.

Authors :
Hun-Jung Cho
Heng Huang
Source :
AIP Conference Proceedings. 8/13/2009, Vol. 1148 Issue 1, p445-448. 4p. 3 Graphs.
Publication Year :
2009

Abstract

For a side-looking frequency-modulated continuous wave (FMCW) radar detector, the information of vehicle speed can not be obtained by Doppler frequency directly. This is due to the detector is perpendicular to marching direction vehicle, therefore, Doppler shift of the beat signal is unapparent. Consequently, the length of detection zone, vehicle length, and the on-time duration of vehicle are three parameters in estimating individual vehicle speed by FMCW radar detectors. Among these parameters, vehicle length and detector on-time can be acquired by time-frequency signal analysis and statistical pattern recognition techniques. Therefore, how to determine the length of detection zone is the remaining problem. In this study, a method of estimating the length of detection zone based on antenna pattern and radar equation is proposed. According to the antenna pattern, the antenna gain is a function of angles in azimuth ([lowercase_phi_synonym]), and elevation (θ) planes, such as G([lowercase_phi_synonym], θ). For each point on the road surface, based on the radar equation, the relative received power can be calculated as long as the position and angle of the detector are given. After setting a threshold value of the relative received power, the length of detection zone on each lane can be determined. In field test, a side-looking 10.5 GHz FMCW radar is used to detect vehicles and the antenna pattern is given. The radar detector is installed at 6 meters height and the angle of depression is nearly 20°. Based on the method described above and set the threshold value equal to 0.0015, then the length of detection zone on each lane can be obtained. According to the length of detection zone, vehicle length, and the on-time duration, the speed of each vehicle can be estimated. Moreover, the accuracy of lane 1 and lane 2 are 84.46% and 83.48% respectively. [ABSTRACT FROM AUTHOR]

Details

Language :
English
ISSN :
0094243X
Volume :
1148
Issue :
1
Database :
Academic Search Index
Journal :
AIP Conference Proceedings
Publication Type :
Conference
Accession number :
43887822
Full Text :
https://doi.org/10.1063/1.3225343