机场跑道异物检测技术研究与实现

发布时间：2018-01-12 23:35

本文关键词：机场跑道异物检测技术研究与实现　出处：《电子科技大学》2014年硕士论文　论文类型：学位论文

【摘要】：机场跑道异物指的是不属于飞机跑道工作区域但又会因各种原因而出现在跑道上的一切“小物体”,一般简称为FOD。FOD会带来严重的安全隐患和经济损失,因此对FOD的检测具有重要的研究意义和使用价值。国外已成功研制出几款不同的检测设备,主要采用雷达检测、视频图像、可见光等技术,国内虽未研制出成品,但也在积极参与探究。本文以机场跑道异物检测为研究背景,首先,基于特定的雷达体制和系统参数仔细分析了雷达监测系统的架构和布局方式,计算了探测范围;其次,根据回波信号模型和杂波模型,讨论了检测前的两种预处理技术,以尽可能降低信杂比,提高检测概率;最后,提出了多目标背景下和非高斯杂波背景下的恒虚警检测技术,实现自动检测。主要工作和创新如下:(1)结合机场工作环境,提出了双侧线性部署的雷达网结构;根据特定的俯仰角度、雷达架设高度、信号带宽等参数,详细分析了雷达网的探测范围和检测精度等;(2)采用毫米波线性调频连续波雷达体制,合理建立回波信号模型;去斜处理是将部分耦合的发射信号作为本振信号,与接收信号混频后取差拍信号,从而获取目标信息;基于维纳滤波的自适应杂波抑制技术是以本地杂波采集数据作为参考信号以预测杂波信号,并与实际采集数据作差值处理,从而提高了信杂比,增大了检测概率;(3)传统的空域恒虚警检测技术已难以在杂波随空域剧烈变化的杂波环境中有效检测到目标,本文提出了基于杂波图恒虚警处理的自动检测算法。剔除平均类杂波图CFAR将已被剔除掉若干个较大样本值后的参考滑窗样本作为杂波功率估计值,从而可以克服多目标的干扰,在强目标干扰下有效地检测到弱目标信号;双参数杂波图CFAR技术根据参考滑窗中的样本均值和样本方差估计背景杂波功率,以实现在韦布尔、对数正态杂波等非高斯背景下检测到目标。通过理论分析和仿真结果验证,所提出的算法均有较好的检测性能。
[Abstract]:A foreign body on an airport runway refers to all "small objects" that do not belong to the runway working area but will appear on the runway for a variety of reasons. Generally referred to as FOD.FOD will bring serious security risks and economic losses, so the detection of FOD has important research significance and use value. Foreign countries have successfully developed several different testing equipment. Radar detection, video images, visible light and other technologies are mainly used. Although no finished products have been developed in China, they are also actively participating in the research. This paper takes the detection of foreign bodies on the airport runway as the research background, first of all. Based on the specific radar system and system parameters, the structure and layout of radar monitoring system are analyzed carefully, and the detection range is calculated. Secondly, according to the echo signal model and the clutter model, two pre-processing techniques are discussed in order to reduce the signal-to-clutter ratio and improve the detection probability as much as possible. Finally, the CFAR detection technology in multi-target background and non-#china_person0# clutter background is proposed to realize automatic detection. The main work and innovation are as follows: 1) combined with airport working environment. A radar network structure with bilaterally linear deployment is proposed. According to the specific pitch angle, radar elevation, signal bandwidth and other parameters, the detection range and detection accuracy of radar network are analyzed in detail. (2) adopting the millimeter-wave linear frequency modulation continuous wave radar system, establishing the echo signal model reasonably; The de-skew processing takes the partial coupling transmission signal as the local oscillator signal, and takes the beat signal after mixing with the received signal, so as to obtain the target information. Adaptive clutter suppression technology based on Wiener filter uses local clutter acquisition data as reference signal to predict clutter signal, and makes difference processing from actual acquisition data, thus improving the signal-to-clutter ratio. The detection probability is increased; 3) the traditional spatial CFAR detection technique has been difficult to detect the target effectively in the clutter environment where the clutter changes sharply with the spatial domain. In this paper, an automatic detection algorithm based on CFAR processing of clutter graph is proposed. The reference sliding window sample which has been removed from several large sample values is taken as the estimated value of clutter power by removing the average clutter graph CFAR. Therefore, the multi-target jamming can be overcome and weak target signal can be detected effectively under strong target jamming. Two-parameter clutter graph CFAR technique estimates background clutter power according to the sample mean and sample variance in the reference sliding window to realize in Weibull. The target is detected under the background of non-#china_person0# such as logarithmic normal clutter. The theoretical analysis and simulation results show that the proposed algorithm has good detection performance.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：V351;TN957.51

【参考文献】