【作者】 张宇；

【导师】 王建新；

【作者基本信息】 南京理工大学 ， 信息与通信工程， 2011， 博士

【摘要】 受多输入多输出(MIMO)通信和综合脉冲孔径雷达(SIAR)的启发,MIMO雷达得到了雷达界的广泛关注和研究。作为一种新体制雷达,其每个天线可以发射任意波形,将空间分集和波形分集的思想引入到雷达中。这使得MIMO雷达具有良好的干扰抵消能力和改善的参数识别能力,可以进行复杂的发射波束形成,可以改善杂波背景中探测低速目标和弱目标的能力,并在低截获以及目标识别等方面比传统雷达有明显优势。本文主要围绕MIMO雷达波形设计和信号处理算法展开相关研究,主要包括以下内容：[1]研究了MIMO雷达中发射波形集设计问题。针对离散频率编码波形(DFCW),分析了其模糊函数的特点。对于满码集DFCW,分析得知其距离自相关函数旁瓣峰值近似为固定值,因此可以在优化设计中只考虑互相关函数项而不考虑自相关函数项,在此基础上提出了低复杂度的优化设计方法；对于非满码集,分析得知其距离自相关函数旁瓣峰值可以低于满码集,在此基础上提出了基于最小最大准则的优化设计方法,实现了具有更好相关性能的波形集设计,并结合速度距离分辨率对两种码集进行了分析。将正交频分复用(OFDM)调制方式引入MIMO雷达中,研究了OFDM编码信号集的优化设计方法。[2]研究了收发共置MIMO雷达接收波束形成技术,从数学角度分析了标准线性阵MIMO雷达阵列排布与波束方向图的关系,针对高旁瓣问题,研究采用基于带零点调向的最小二乘方向图合成方法进行抑制。针对MIMO雷达接收端可以形成虚拟孔径的特点,将传统的常规波束形成和最小方差无畸变响应(MVDR)波束形成方法推广应用到MIMO雷达中,并与其在传统相控阵雷达下的性能进行了分析比较。针对采样矩阵求逆法(SMI)波束形成器的稳健性,研究了MIMO雷达体制下对角加载技术的应用。[3]研究了收发共置MIMO雷达发射波束形成技术。MIMO雷达在发射信号选择上的灵活性使其可以进行复杂的发射波束方向图合成,其方法主要是通过改变发射信号之间的互相关性来进行不同形状的发射波束方向图设计。基于发射信号互相关矩阵设计的思想,研究了在最小二乘和最小最大准则下发射波束形成的设计方法,分别基于梯度搜索法和遗传算法建立优化模型。针对发射方向图波形设计问题,提出基于已知互相关矩阵和期望方向图两种波形设计方法。[4]针对MIMO雷达接收端信号分离问题,研究了匹配滤波方法和非匹配滤波方法的应用。针对基于匹配滤波器进行信号分离的方法得到的主旁瓣比较差的特点,引入基于复合滤波进行信号分离的方法和基于MMSE准则进行信号分离的方法进行改进。两种方法均实现了对分离输出信号主旁瓣比的改善。基于复合滤波进行信号分离的方法可以调整脉冲压缩后的信号主瓣宽度,有利于提高分辨率,但其信噪比损失较大；基于MMSE准则的方法信噪比损失较小,且对多普勒有较好的适应性。更多还原

【Abstract】 Inspired by multiple-input multiple-output (MIMO) communication technique and synthetic impulse and aperture radar (SIAR), MIMO radar is proposed and becomes the research focus concerned by scholars from many countries. As a new radar, MIMO radar is capable of transmitting arbitrary waveform from each antenna element, so it can exploit waveform diversity and space diversity. By exploiting these potentials, MIMO radar has several advantages, such as excellent interference rejection, improved parameter identifiability, enhanced flexibility for transmit beampattern design, advanced ability of low speed target detection and weak target detection in clutter environment, improved performance on low probability of intercept and target recognition.In this dissertation, MIMO radar waveform design and signal processing algorithms are investigated, and the main research focus on the following issues:[1] Investigate the design method of transmitting waveforms for MIMO radar. Based on the analysis of the ambiguity function, the performance of discrete frequency coding waveform (DFCW) is studied. For full-code set DFCW, the normalized range autocorrelation sidelobe peak almost does not depend on the firing order of frequency, so waveform design method with low complexity is proposed which can only consider the range cross-correlation to design the DFCW full code set; for non-full code set DFCW, its bound of ASP is lower than full code set’s, so waveform design method with better correlation performance based on minimax criteria is proposed, and then range resolution and velocity resolution are analyzed. Orthogonal frequency division multiplexity (OFDM) used in MIMO radar is analyzed, and OFDM-code waveform design method is proposed.[2] Investigate receive beamforming of colocated MIMO radar. Based on mathematical analysis, the connection of antenna spacing and beampattern for MIMO radar with standard linear array is given. For grating lobes, a novel algorithm which can synthesize least squares error (LSE) receiving pattern subject to a set of null constraints is proposed. For additional virtual sensors at receiver of MIMO radar, conventional beamformer and minimum variance distortionless response (MVDR) beamformer are extended and applied for MIMO radar. And the performance of beamformers is compared under MIMO radar and phased array radar. For improving the robustness of sample matrix invert algorithm, diagonal loading is applied under MIMO radar.[3] Investigate the transmit beamforming technology of MIMO radar. MIMO radar can synthesize a desired spatial transmit beampattern through the selection of signal sets with arbitrary cross-correlation properties. For a given transmit beampattern, it can be achieved by designing the cross-correlation matrix of the probing signal vector transmitted by MIMO radar. Using the designing idea, the methods based on least square (LS) criteria and minimax criteria are studied, and the optimizing models based on gradient search and genetic algorithms are built. For designing a signal set which can approximate a desired spatial transmit beampattern, the methods based on known cross-correlation matrix and desired spatial transmit beampattern are proposed respectively.[4] For separating the echoes of different transmitting waveforms of MIMO radar, the methods based on matched filter and mismatched filer are studied. Because the performance of the separating method based on matched filter is poor, it can be improved by using the idea of compound filter and the approach based on a minimum mean-square error (MMSE) formulation. Based on the idea of compound filter containing a pre-filter and an inverse filter, a novel method of separating the echoes of different transmitting waveforms of MIMO radar is proposed, but the loss of signal to noise ratio (SNR) is large. The approach based on a MMSE formulation exhibits small SNR loss and is robust to doppler mismatch.更多还原