【作者】 张劲东；

【导师】 朱晓华；

【作者基本信息】 南京理工大学 ， 信息与通信工程， 2010， 博士

【摘要】 波形分集是一种重要的分集方式,已成为现代雷达系统发展的重要方向之一。在这种新分集方式下,雷达根据环境实时地改变自身发射波形,可自适应地优化检测、跟踪和抗干扰等方面的性能,且使敌方的电子侦察及干扰更为困难。在自适应雷达系统中波形分集技术的研究对于探求自适应雷达的新理论、新方法,提高雷达在未来战场上的性能和抗干扰能力具有重要的理论意义和实用价值。本文系统地研究了自适应雷达系统中波形分集技术的理论和方法,展开的主要工作包括：[1]针对交替搜索与检测的自适应工作方式,系统分析了最优匹配照射—接收机的基本原理,基于相位编码雷达信号着重展开了目标多散射中心的提取和最优匹配照射—接收机的设计方法研究；考虑到递归得到的目标多散射中心距离像信息可作为下一次计算的先验信息,提出了基于递归最大信噪比准则的自适应距离旁瓣抑制方法,可稳健地实现目标各个散射中心位置的提取和复幅度的高精度估计,为目标多散射中心距离像的精准模型建立奠定了基础；提出了基于相位域共轭梯度算法的匹配照射相位编码信号设计方法,具有收敛速度快、运算量小的特点,设计得到的相位编码信号的性能接近最优匹配照射信号；研究结果表明,匹配照射不但提高了匹配滤波器的输出信噪比,而且降低了在不同目标姿态角下输出信噪比的起伏,因此具有匹配照射功能的自适应雷达可获得更为良好和平稳的检测性能。该结论对未来雷达系统的设计具有一定的借鉴意义。[2]针对有源干扰,尤其是频域窄带干扰和欺骗式干扰,基于波形分集技术进行了自适应雷达的干扰对抗研究。在已知其它多种电子系统的干扰频谱先验信息的前提下,提出了一种用于叶簇穿透超宽带雷达在干扰频带形成自适应陷波的相位编码信号设计及频域处理方法,设计所得信号具有近似理想功率谱的同时通过频域失配处理获得良好的距离分辨力和主旁瓣比；针对距离波门拖引的欺骗式干扰下的目标检测,充分利用干扰机的复制能力进行反欺骗,提出了具有频域正交性的发射波形构造和相应的处理方法,在消除干扰的同时利用波形分集能量的累积提高了距离欺骗式干扰下当前目标的跟踪和多目标的检测能力；在脉冲多普勒雷达在速度欺骗式干扰下,设计了一种新的对抗思路,通过脉问自适应调相使干扰假目标信号在被拖引目标多普勒范围形成保护凹口,从而不影响当前目标的检测,给出了脉间自适应调相的设计方法,提出了基于多通道相关处理的干扰机模式及参数提取方法,同时设计了欺骗式干扰下自适应对抗的工作流程,实现了大功率速度欺骗式干扰下的雷达自主对抗。[3]针对自适应多输入多输出(MIMO)雷达,采用模糊函数对收发共置和双基地两种形式的MIMO雷达的波形优化进行了研究。给出了收发共置MIMO雷达距离和角度分辨力,分析指出其匹配滤波器的峰值受到目标真实空域频率与假设空域频率之差的影响,输出旁瓣受到目标真实空域频率和假设空域频率共同影响,根据收发共置MIMO雷达的空域分辨力对其探测空域进行划分,提出了基于空域频率变化的收发共置自适应MIMO雷达正交波形设计方法,降低了匹配滤波器输出的峰值旁瓣电平；在研究双基地MIMO雷达距离速度模糊函数的基础上,推导了双基地MIMO雷达的距离和速度分辨力,提出了基于目标空间位置的双基地自适应MIMO雷达正交波形设计方法,通过频率编码信号码元宽度、码元个数与频率编码的联合优化设计,双基地MIMO频率编码雷达信号在保持所需距离速度分辨力的同时使相关输出的积分旁瓣电平最小化,克服了双基地MIMO雷达距离速度模糊函数随双基地角增大而分辨力下降的问题,该方法同样适用于多基地和组网雷达系统的性能优化。[4]针对MIMO波束形成技术,采用现代优化理论对MIMO雷达发射波束形成和匹配滤波波束形成技术进行了研究。将MIMO雷达的自适应发射方向图设计问题转化为半正定二次锥规划问题,提出了基于现代凸优化理论的发射互相关矩阵优化方法,使设计得到的方向图具有期望的主瓣形状和指定角度区域的最小化副瓣电平；针对MIMO雷达接收端通过匹配滤波波束形成进行数字多波束设计的问题,提出了一种新的匹配滤波波束形成模型,指出该模型中对MIMO雷达的发射端和接收端进行协同设计,即发射信号和波束形成滤波器的联合优化,可大大扩展MIMO雷达系统的自由度,给出了在MIMO雷达接收端进行数字多波束形成时发射信号和波束形成滤波器优化设计的目标函数和相应的优化方法,获得了比已有匹配滤波波束形成模型更好的方向图性能。更多还原

【Abstract】 As an important diversity, waveform diversity (WD) technique has been an important development direction of the modern radar system. With this new diversity technique, the performance of detection, tracking and anti-jamming of radar can be optimized by altering its transmitted waveform, and at the same time it’s difficult for enemy to perform electronic reconnaissance and jamming. The theory investigation of WD in adaptive radar system is of much theoretical and applicable significance in the exploitation of adaptive radar system and the improvement of the performance and anti-interference ability in future battlefield.In this thesis the theory and method of WD in adaptive radar system are investigated and main works are accomplished as follows:Firstly, basic principle of optimum matched illumination and reception is analyzed for the adaptive operation mode of alternative search and detection for radar. Extraction of multiple scattering points of the target and design method of optimum matched illumination transmitted waveform and the correspondence receiver are investigated based on phase-coded radar signal. It is considered that the apriori information obtained by recursive function can be used to adaptively suppress range sidelobes. An adaptive range sidelobe suppression algorithm based on recursive maximum signal-to-noise ratio (SNR) criteria is presented. Every scattering point of the target can be extracted reliably and complex amplitude of the scattering point can be accurately estimated. The accuate target model with multiple scattering points can be constructed. The matched illumination phase-coded signal design method based on conjugate gradient algorithm in phase domain is also investigated. The phase-coded radar signal derived by this algorithm approaches the optimum matched illumination signal and this algorithm enjoys fast convergence and small computation amount. The results show that matched illumination not only increases the SNR of the matched filter output, but also decreases the variation of the output SNR at different attitude of the target. Better and more stable detection performance can be obtained by adaptive radar system and this conclusion can be used for adaptive radar system design in future.Secondly, anti-jamming technique based on WD for countering active jamming, especially the narrow-band interference (NBI) and deception jamming, is studied. With the apioror information of the NBIs from other electronic systems, a new kind of phase-coded signal design method which minimizes the power spectrum in bands of the interference and the correspondence frequency domain processing method are proposed for foliage penetration ultra wide band radar (FOPEN UWBR). The synthesized signal has the approximatly ideal power spectrum and good range resolution and high peak sidelobe ratio (PSLR) after frequency domain mismatch processing. For target detection under deception jamming of range gate pull off (RGPO), a set of specified transmitted signals orthogonal in frequency and the correspondence processing method are presented to eliminate the jammer signal which is a copy of the transmitted radar signal. This method improves the target tracking and multiple target detection performance by WD. For target detection under synchronous range and velocity jamming, a novel anti-jamming method of removing deceiving decoys based on transmitted signals with adaptive random initial phases is studied. Adaptive random initial phases which can form a notch covering the Doppler frequency of the target make detection unaffected. The multi-channel auto correlation processing method is given to extract the manner the jammer operates and the parameters of the false targets. Work flow of a pulse Doppler radar is also designed. The simulation results show the anti-jaming capability of radar by itself for countering high power jamming.Thirdly, to optimize adaptive multiple-input multiple output (MIMO) radar performance, colocated and bistatic MIMO radar are studied based on ambiguity function. Range and spatial resolution of the collocated MIMO radar is given. It is shown that the peak value of the matched filter is affected by the difference of the real target and assumed spatial frequency and the sidelobe is determined by the real target and assumed spatial frequency. According to this property, an algorithm for adaptively designing orthogonal frequency-hopping waveforms according to the target spatial frequency is proposed. The search spatial region is divided into multiple sections by spatial resolution and waveforms are optimized to improve the peak sidelobe level (PSL) of the matched filter output for for the collocated MIMO radar. Characteristic of range-velocity ambiguity functioin of the bistatic MIMO radar is also analyzed, and range and velocity resolution of bistatic MIMO radar are deduced. An algorithm for adaptively designing orthogonal waveforms according to the target position is proposed. This algorithm jointly optimize the code width, the number of the code and code set of the frequency code waveforms and improves the range and velocity resolution and the integrated sidelobe level (ISL) of the matched filter output at the correspondence position. This method sloves the problem that the range and velocity resolution decreases with the increased bistatic angle, and can also be applied in performance optimization for multistatic and netted radar systems.Finally, transmit beamforming and matched filter beamforming are studied for MIMO radar based on modern optimization theory. The adaptive transmit beampattern design for MIMO radar is converted to a semi definite second order cone programming model. The cross-correlation matrix optimization method based on modern convex optimization theory is proposed. Simulation results show the synthesized beampattern can keep the main beam shape and minimize the sidelobe in specified angle section. Digitial multi-beam forming in MIMO radar receiver is also studied. A new matched filter beamforming model is proposed. In this model, the transmitter and receiver can be designed simultaneously by jointly optimizing the cross-correlation matrix of the transmitted signals and the beamformer, therefore greatly increasing the freedom degrees of MIMO radar system. The penalty fuction for multi-beam forming optimization design is built and the correspondence optimization method is given. Simulation results show that this method can obtain better beampattern than that derived by the original matched filter beamforming model.更多还原