基于特征量分析的距离扩展目标检测技术
发布时间：2018-07-15 16:39
【摘要】：宽带雷达在距离分辨力、目标参数测量精度、雷达的“四抗”性能、目标的识别性能、强杂波背景下弱目标的检测性能、信号的波形设计等方面相较于窄带雷达有明显的优势,是现代雷达检测技术的重要方向之一。随着各种新技术的兴起,对雷达的分辨力的要求也越来越严格。当雷达距离分辨力比目标尺寸小很多的时候,目标表现为距离扩展目标特性,传统的点目标检测不能充分利用其回波特性,不在适用于目标信号的检测。因此,亟需对高分辨力雷达距离扩展目标进行研究。本文围绕着距离扩展目标的特征量检测问题,开展了如下工作:1、介绍了宽带雷达方程,并对距离扩展目标的RCS统计模型进行分析。2、研究了距离扩展目标常用的LFM发射波形与脉冲压缩技术,并对距离扩展目标回波特性进行仿真分析3、介绍了几种经典的杂波分布模型,分析了广义复合分布杂波模型在不同参数的情况下可以退化成几种经典的杂波分布模型,同时用ZMNL方法对其进行仿真验证。4、介绍了三种现有的距离扩展目标检测算法:能量积累检测算法,最佳匹配检测算法和M/N检测算法。并用蒙特卡洛实验法仿真分析其算法优劣性,并在不同的条件下对比其检测性能。5、改进的距离扩展目标检测算法:偏斜度特征量和尖峰度特征量。根据目标回波和杂波拖尾的厚度和高峰的尖峭程度,对其偏度或者峰度进行检测,并用蒙特卡洛实验法仿真分析其算法优劣性,并在不同的条件下对比其检测性能。6、改进的距离扩展目标检测算法:三阶累积量特征量。根据三阶累积量能够自动消除高斯背景噪声的特性,对其三阶累积量进行检测,并用蒙特卡洛实验法仿真分析其算法优劣性,并在不同的条件下对比其检测性能。本篇文章第一步介绍了距离扩展目标检测的相关理论,接下来研究了适用于距离扩展目标的检测的相关模型和距离扩展目标的杂波模型,通过仿真验证了距离扩展目标经典的检测算法,然后提出了基于特征量分析的距离扩展目标检测新算法,最后通过仿真分析对比新算法的优劣性。
[Abstract]:Wideband radar has obvious advantages over narrowband radar in range resolution, target parameter measurement accuracy, radar's "four resistance" performance, target recognition performance, weak target detection performance in strong clutter background, signal waveform design, etc. It is one of the important directions of modern radar detection technology. With the rise of various new technologies, the requirement of radar resolution is becoming more and more strict. When the range resolution of radar is much smaller than that of the target, the target is characterized by extended range. The traditional point target detection can not make full use of its echo characteristics and is not suitable for target signal detection. Therefore, it is urgent to study the range extended target of high resolution radar. In this paper, we focus on the feature detection of extended range targets, and introduce the wideband radar equation as follows: 1. The RCS statistical model of the extended range target is analyzed, and the LFM transmitting waveform and pulse compression technology are studied. The echo characteristics of the extended range target are simulated and analyzed. 3. Several classical clutter distribution models are introduced. The generalized composite distributed clutter model can degenerate into several classical clutter distribution models under different parameters. At the same time, the ZMNL method is used to verify the algorithm. 4. Three existing distance extended target detection algorithms are introduced: energy accumulation detection algorithm, best match detection algorithm and M / N detection algorithm. The Monte Carlo experiment method is used to simulate and analyze the advantages and disadvantages of the algorithm, and its detection performance is compared under different conditions. The improved distance extended target detection algorithm includes skewness characteristic quantity and spike degree characteristic quantity. According to the thickness of target echo and clutter tail and the sharp degree of peak, the skewness or kurtosis of target echo and clutter tail is detected, and the algorithm is simulated and analyzed by Monte Carlo experiment. At the same time, the detection performance is compared under different conditions. 6, and the improved distance extended target detection algorithm: the third order cumulant feature. According to the characteristic that third-order cumulant can automatically eliminate the background noise of Gao Si, the third-order cumulant is detected, and its algorithm is simulated and analyzed by Monte Carlo experiment, and its detection performance is compared under different conditions. In the first step of this paper, we introduce the theory of range-extended target detection. Then we study the model of range-extended target detection and the clutter model of distance-extended target. The classical distance extended target detection algorithm is verified by simulation. Then a new distance extended target detection algorithm based on feature analysis is proposed. Finally, the advantages and disadvantages of the new algorithm are compared by simulation.
【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】：TN957.51
本文编号：2124712
[Abstract]:Wideband radar has obvious advantages over narrowband radar in range resolution, target parameter measurement accuracy, radar's "four resistance" performance, target recognition performance, weak target detection performance in strong clutter background, signal waveform design, etc. It is one of the important directions of modern radar detection technology. With the rise of various new technologies, the requirement of radar resolution is becoming more and more strict. When the range resolution of radar is much smaller than that of the target, the target is characterized by extended range. The traditional point target detection can not make full use of its echo characteristics and is not suitable for target signal detection. Therefore, it is urgent to study the range extended target of high resolution radar. In this paper, we focus on the feature detection of extended range targets, and introduce the wideband radar equation as follows: 1. The RCS statistical model of the extended range target is analyzed, and the LFM transmitting waveform and pulse compression technology are studied. The echo characteristics of the extended range target are simulated and analyzed. 3. Several classical clutter distribution models are introduced. The generalized composite distributed clutter model can degenerate into several classical clutter distribution models under different parameters. At the same time, the ZMNL method is used to verify the algorithm. 4. Three existing distance extended target detection algorithms are introduced: energy accumulation detection algorithm, best match detection algorithm and M / N detection algorithm. The Monte Carlo experiment method is used to simulate and analyze the advantages and disadvantages of the algorithm, and its detection performance is compared under different conditions. The improved distance extended target detection algorithm includes skewness characteristic quantity and spike degree characteristic quantity. According to the thickness of target echo and clutter tail and the sharp degree of peak, the skewness or kurtosis of target echo and clutter tail is detected, and the algorithm is simulated and analyzed by Monte Carlo experiment. At the same time, the detection performance is compared under different conditions. 6, and the improved distance extended target detection algorithm: the third order cumulant feature. According to the characteristic that third-order cumulant can automatically eliminate the background noise of Gao Si, the third-order cumulant is detected, and its algorithm is simulated and analyzed by Monte Carlo experiment, and its detection performance is compared under different conditions. In the first step of this paper, we introduce the theory of range-extended target detection. Then we study the model of range-extended target detection and the clutter model of distance-extended target. The classical distance extended target detection algorithm is verified by simulation. Then a new distance extended target detection algorithm based on feature analysis is proposed. Finally, the advantages and disadvantages of the new algorithm are compared by simulation.
【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】：TN957.51
【参考文献】
相关期刊论文 前2条
1 ;Two adaptive detectors for range-spread targets in non-Gaussian clutter[J];Science China(Information Sciences);2011年02期
2 陆林根;高距离分辨(HRR)雷达单个目标回波信号检测[J];系统工程与电子技术;1999年09期
相关硕士学位论文 前1条
1 张广宇;基于统计特性的宽带雷达信号检测方法研究[D];电子科技大学;2010年
，本文编号：2124712
本文链接：https://www.wllwen.com/kejilunwen/wltx/2124712.html
