中文摘要：

针对传统的优效粒子滤波的检测前跟踪算法（EPF—TBD）在低信噪比场景中目标状态变化会引起虚警和漏检问题，文中给出一种改进的优效PF—TBD算法改进其检测方式，利用目标检测概率变化的斜率和当前目标状态进行检测。仿真结果表明，与传统算法相比，改进算法检测效果有明显的改善，平均检测概率提高，平均虚警概率降低，通过对比不同门限的检测效果给出适应不同要求的改进算法检测门限。

英文摘要：

The traditional Efficient PF- TBD （EPF -TBD, PF -TBD： TBD algorithm based on the particle filter） algorithm suffers the leak detection and false alarm in case of state changes of targets with low SNR. This article puts forward an improved EPF -TBD algorithm improved for better detecting probability of slope and the current target state for testing. The simulation results show that compared with the traditional EPF -TBD algorithm, the improved EPF -TBD algorithm is more effective in relieving the leak detection and false alarm problem with better detection performance. The detection thresholds of the improved algorithm for different requirements are given by comparing the different threshold detection effect.