中文摘要：

针对电磁探测卫星的重访时间间隔长、且捕获得到目标信息的时间为随机,目标运动模型难于精确建立,数据杂波干扰强等问题,结合卫星电子信息给出的辐射源电磁特征,基于粒子滤波提出了一种稳健的海上舰船目标跟踪算法。首先,采用二阶自回归的状态转移模型确定候选量测的关联区域,从而减少杂波的干扰;然后选取与目标电磁特征相似的量测进行粒子滤波的状态更新,并通过重采样操作剔除权值较小的粒子以提高跟踪算法的精确性与稳健性。仿真与真实数据的实验结果表明,该方法在强杂波干扰下可以稳定跟踪卫星电子信息中的海上舰船目标并且具有较高的精确性与稳健性。

英文摘要：

Aim at the issues of long recurrence interval,random time of the data,strong clutter interference,and the difficulties of modeling target＇s motion in the application of electronic reconnaissance satellite,a novel ship target tracking algorithm based on particle filtering for satellite electronic information is proposed in this paper to solve these ship target surveillance problems. Firstly,the association area of the measurements is selected by using the auto-regressive state transition model to suppress the clutter. Secondly,the measurements with similar electromagnetic characteristics to the ship target are utilized to update the weights of the particles. Finally,resampling is considered to remove the particles with small weights and to improve the tracking accuracy. Experimental results on both simulated and real world data demonstrate that the proposed algorithm can stably track the ship targets under strong clutter interference and has substantial improvements in terms of accuracy and robustness.