Abstract: In the out-field experiment,the amount of echo is huge based on wideb and direct sampling model. In addition, the scattering point migration of ballistic target results in serious micro motion and dramatically spectrum change. In such case, the traditional motion compensation methods are unsuitable anymore and it is difficult in processing wide band direct sampling data. To tackle such problems, an accurate algorithm for wide band high velocity ballistic targets motion compensation is proposed. Firstly, down-sample rate is performed on the wide band data to reduce the data amount and group targets are separated by region Hough transform on the desampled data.Then the target trajectories can be separated.The corresponding location in the origin wideband data can be computed. Next, signal separation is executed after preliminary motion compensation. Further compensation is made upon the extracted signal by gravity center method. The residual motion parameters and smooth micro-Doppler (m-D) feature curve can be obtained by performing time-frequency analysis based on the time varying autoregressive model. The m-D parameters are used to compensate the target signal accurately. Finally, CADFEKO simulation results verify the effective ness of the proposed method.

Key words: micro motion, high velocity group targets, ballistic targets motion compensation, region Hough transform, time-frequency analysis, time varying autoregressive model