Moving vehicle classification based on micro-Doppler signature

Micro-Doppler information contained in the returned echo reflects the special structure of moving vehicles and can be utilized to discriminate wheeled vehicle and tracked vehicle. In this paper, generalized matched filter is employed to suppress clutter and a preprocessing is used to eliminate the influence of bulk velocity for Doppler spectrum firstly. Then features which depict the distinction between the wheeled and tracked vehicles are extracted and experimental results based on extracted features are presented.