中文摘要：

针对电子侦察卫星的重访时间较长且随机、舰船目标运动模型难于精确建立、数据杂波干扰强、多辐射源扩展目标跟踪等问题，结合卫星得到的辐射源特征参数提出了基于多假设跟踪（multiple hypothesis tracking，MHT）改进的舰船目标跟踪算法。首先分析了卫星电子信息舰船目标跟踪的特点；在没有辐射源类别与个数等先验知识的情况下，利用辐射源位置和载频等信息进行了两次聚类，实现了数据压缩以及杂波抑制；再在MHT的框架上利用目标运动状态信息结合辐射源的载频特征信息实现了多目标多辐射源的边跟踪边参数估计。仿真对比实验结果表明，结合辐射源特征信息的方法具有更好的跟踪性能，具有较高的跟踪精确性与稳健性。

英文摘要：

Aiming at the problems of the long and random revisit time of electromc reconnaissance satellites, difficulties of establishing ship target＇s motion model accurately, strong clutter interference, extended target tracking of multi-emitters and other issues, a novel ship target tracking algorithm is proposed based on the improved multiple hypothesis tracking （MHT）. Firstly, the characteristics of the satellite electronic infor- mation for ship target tracking are analyzed. Under the condition that there is no prior knowledge of the class and number of the emitters, data compression and clutter removal are achieved after two clustering by using the information of position and carrier frequency of the emitter. And then tracking and parameter estimation of multi-target with multi-emitter are completed by using the information of target motion state combined with carrier frequency characteristic information of the emitter based on the MHT frame. The simulation results show that proposed method has better tracking performance with high tracking accuracy and robustness.