混合估计多模粒子滤波的机动弱目标检测前跟踪算法

发布时间：2018-01-18 11:24

本文关键词：混合估计多模粒子滤波的机动弱目标检测前跟踪算法　出处：《控制与决策》2014年03期 　论文类型：期刊论文

【摘要】：针对低信噪比条件下的机动小目标实时检测与跟踪困难的问题,提出一种基于混合估计多模粒子滤波的检测前跟踪改进算法.首先根据前一时刻所采用的模型状态及其转移概率等先验信息实现当前时刻的模型采样;然后在充分考虑当前量测下实现当前的粒子预测,采用一种序贯重要性平滑重采样策略,在不增加计算量的前提下,改善了粒子多样性衰退的问题;最后通过新的粒子集完成对模型和状态的合理估计与目标检测.仿真结果验证了该方法的检测与跟踪性能优于传统的多模粒子滤波方法.
[Abstract]:It is difficult to detect and track small maneuvering targets in real time under low SNR. An improved pre-detection tracking algorithm based on hybrid estimation multimode particle filter is proposed. Firstly, the model sampling at the current time is realized based on the prior information such as the model state and its transition probability. Then a sequential importance smoothing resampling strategy is adopted to realize the current particle prediction under the full consideration of the current measurement. The problem of particle diversity decay is improved without increasing the computational complexity. Finally, the new particle set is used to estimate the model and state reasonably and to detect the target. The simulation results show that the performance of the proposed method is better than that of the traditional multimode particle filter.
【作者单位】：空军预警学院雷达兵器运用工程军队重点实验室;
【基金】：国家自然科学基金项目(61072132) 中国博士后科学基金项目(2013M542541)
【分类号】：TN953
【正文快照】： 0引言对于远距离高机动的小目标,雷达接收到的回波信号相当微弱,用传统的积累与检测方法不能可靠地检测目标[1-3],而检测前跟踪(TBD)技术则比较适合于低信噪比微弱目标.TBD是将一次扫描回波做脉冲串积累后不做门限比较(或者与低门限比较),从而保留下来微弱的目标信息,在得到多

【参考文献】