基于代价参考粒子滤波的存在概率检测算法

发布时间：2018-06-12 05:55

本文选题：代价参考粒子滤波 + 检测前跟踪　；参考：《电子学报》2017年06期

【摘要】：为提高统计特性未知情况下对非线性微弱动目标的检测能力,本文提出一种基于代价参考粒子滤波的检测前跟踪算法.首先在代价参考粒子滤波的状态向量中增加模拟目标存在状态的离散变量,并在离散变量的转移过程中引入相关系数判决机制;其次,利用代价参考粒子滤波的输出估计存在概率;最后,基于存在概率构造检验统计量.当检验统计量大于给定门限时宣布目标出现.天波雷达目标检测的仿真表明,当系统的统计特性已知时,该方法的检测性能与基于传统粒子滤波的似然比检测、存在概率检测等相当;当统计特性未知时,该方法的检测性能比传统方法提高了2d B以上.本方法可用于复杂背景下的监测系统,如雷达、声呐等.
[Abstract]:In order to improve the detection ability of nonlinear weak moving targets with unknown statistical characteristics, a pre-detection tracking algorithm based on cost reference particle filter is proposed in this paper. Firstly, the discrete variables of the existing state of the target are added to the state vector of the cost reference particle filter, and the correlation coefficient decision mechanism is introduced in the transfer process of the discrete variables. The existence probability is estimated by using the cost reference particle filter. Finally, the test statistics are constructed based on the existence probability. A target is declared when the test statistic is greater than a given threshold. The simulation of sky-wave radar target detection shows that when the statistical characteristics of the system are known, the detection performance of this method is equivalent to that of the likelihood ratio detection based on traditional particle filter. The detection performance of this method is more than 2 dB higher than that of the traditional method. This method can be used for monitoring systems in complex background, such as radar, sonar and so on.
【作者单位】：陕西科技大学电气与信息工程学院;西安工程大学电子信息学院;
【基金】：陕西科技大学启动基金(No.BJ15-10,BJ02) 西安市科技计划项目(CXY1513) 国家自然科学基金(No.61401347)
【分类号】：TN713;TN957.51

【相似文献】