岑宗骏,
汤俊,
清华大学电子工程系 北京 100084
基金项目:国家部委专项项目(19-163-11-ZD-019-006-02)
详细信息
作者简介:李波：李 波(1996–)，男，广西陆川人，清华大学电子工程系在读硕士生，研究方向为外辐射源信号处理、弱目标检测等。E-mail: lib17@mails.tsinghua.edu.cn
岑宗骏(1995–)，男，广西北流人，清华大学电子工程系在读博士生，研究方向为阵列信号处理、软件化雷达等。E-mail: cenzhongjun123@163.com
汤俊：汤 俊(1973–)，男，江苏南京人，博士，教授，2000年在清华大学电子工程系获得博士学位，现为清华大学电子工程系教授，研究方向为阵列信号处理、软件化雷达等，目前发表文章百余篇。E-mail: tangj_ee@tsinghua.edu.cn
通讯作者:汤俊 tangj_ee@tsinghua.edu.cn
责任主编：万显荣 Corresponding Editor: WAN Xianrong中图分类号:TN958
计量
文章访问数:1707
HTML全文浏览量:458
PDF下载量:152
被引次数:0
出版历程
收稿日期:2020-03-23
修回日期:2020-05-22
网络出版日期:2020-06-18
A New Method of Target Detection for Passive Radar Based on Information Accumulation
LI Bo,CEN Zongjun,
TANG Jun,
Department of Electronic Engineering, Tsinghua University, Beijing 100084, China
Funds:The National Ministry Foundation of China (19-163-11-ZD-019-006-02)
More Information
Corresponding author:TANG Jun, tangj_ee@tsinghua.edu.cn
摘要
摘要:外辐射源雷达系统反隐身性能强、隐蔽性好、生存能力强，在军用和民用领域都具有十分广阔的应用场景。为了有效地对低信噪比的弱目标进行检测，并且同时满足系统的实时性需求，该文针对外辐射源雷达系统的特点，依据检测前跟踪算法的思想，提出一种基于信息积累的外辐射源雷达系统目标检测方法。该方法首先将目标状态空间离散格点化，然后利用递推贝叶斯滤波的思想在多帧观测数据之间进行目标状态信息的传递和积累，最后利用信息熵作为判决目标是否存在的条件，避免了对目标存在和目标不存在两种状态之间转移概率模型的先验假设，是一种实现简单、计算复杂度低、可并行度高的目标检测方法。实验结果表明，该方法不仅运行时间短，实时性能强，而且具有良好的检测性能和一定的鲁棒性。
关键词:外辐射源/
目标检测/
检测前跟踪/
贝叶斯滤波/
信息熵
Abstract:Owing to their strong anti-stealth performance, good concealment and strong survivability, passive radar systems have a wide range of applications in both military and civilian fields. We propose a method of target detection for passive radar systems which is based on the characteristics of these systems and the track-before-detect concept. This method accumulates information to effectively detect weak targets with low signal-to-noise ratios and meet real-time requirements. First, we discretize the state space, then perform recursive Bayesian filtering to transfer and accumulate target-state information between multiple frames. Lastly, the information entropy is used to determine whether the target exists, thereby avoiding reliance on a prior assumption about the transition probability model between the existence and the absence of the target. This method is simple to implement and has low computational complexity and high parallelism. The experimental results indicate that the proposed method has a short running time and strong real-time performance, as well as good detection performance and robustness.
Key words:Passive radar/
Target detection/
Track-Before-Detect (TBD)/
Bayesian filtering/
Information entropy
PDF全文下载地址:
https://plugin.sowise.cn/viewpdf/198_fb54171a-972f-4bcb-ad6f-7ba41b7784d1_R20023
