基于M-D特征和Duffing振子的车辆目标分类方法

Classification method for vehicle targets based on Micro-Doppler characteristics and duffing oscillator

针对低信噪比条件下履带式车辆和轮式车辆的分类问题,提出一种基于微多普勒(Micro-Doppler,MD)特征和Duffing振子的低信噪比条件下车辆目标分类方法.首先,基于窄带雷达对两类车辆目标进行回波建模,分析了两类车辆目标微多普勒特征的主要差异,然后利用Duffing振子系统检测微弱信号的性能优势,在低信噪比条件下基于Duffing振子实现了两类车辆目标较高精度的分类,最后基于实测数据验证了所提方法的有效性.

The classification of wheeled vehicles and tracked vehicles with low signal-to-noise ratio（SNR）is a difficulty in the research field of vehicle recognition.For solving the problem,a method based on Micro-Doppler（M-D）characteristics and Duffing oscillator was proposed for vehicle classification in low SNR conditions.Firstly,the echoes model based on narrow band radar was established for two kinds of vehicles.The main difference of Micro-Doppler characteristics was analyzed for the two kinds of vehicles.Then the performance advantage for detecting weak signal of duffing oscillator system was applied in vehicles classification in low SNR conditions.The classification with high accuracy was obtained for the two kinds of vehicles.Finally,the effectiveness of the method was validated based on the measured data.