面向参数估计的认知雷达自适应发射波形优化技术研究

发布时间：2018-05-05 22:34

本文选题：认知雷达 + 自适应　；参考：《国防科学技术大学》2014年博士论文

【摘要】：现有雷达系统的发射波形和工作模式一般固定不变,接收系统仅被动地处理雷达信号和数据,使得即使精心设计的雷达系统也难以在复杂的时空多变环境中获得理想的处理性能。认知雷达具有接收机到发射机的闭环反馈机制,能够根据先验和实时的环境及目标知识自适应动态调整雷达收发系统,从根本上改变了传统雷达单向的信息处理方式,能够大幅提高雷达系统性能和环境适应能力,具有重要的理论研究和军事应用价值,受到国内外很多专家学者和军方的广泛关注。自适应发射波形优化技术是认知雷达系统的基础关键技术之一,论文主要研究了面向目标参数估计问题的雷达发射系统自适应优化调整技术,具体内容可分为目标跟踪(运动目标状态估计)和雷达成像(静止目标电磁散射系数估计)两个方面:在目标跟踪方面,研究了杂波环境下以提高机动目标跟踪精度为目的的自适应波形选择技术。论文首先对雷达接收系统和跟踪测量数据获取过程进行了详细建模,分析了不同信噪比情况下雷达发射波形对测量数据统计特性的影响及定量计算方法,并设计了与自然界中蝙蝠发射声波信号类似的CF-HFM(Constant Frequency-Hyperbolic Frequency Modulation)雷达脉冲串波形,提出了杂波环境下基于雷达任务的机动目标跟踪自适应波形选择算法及自适应发射波形与检测门限联合优化方法;其次,为了较为精确地估计机动目标混合运动状态后验概率密度函数,提出了一种完全分离连续基状态与离散模型状态估计的交互多模型概率数据关联粒子滤波器(Interacting Multiple Model Probabilistic Data Association Particle Filter,IMMPDAPF),估计循环仅需要基状态粒子,能够控制每个运动模型对应的粒子数,并引入稳健性较好的Amariα-散度表征不同发射波形对应雷达测量数据包含的信息量,提出了与IMMPDAPF实现过程紧密结合的杂波环境下基于信息论准则的机动目标跟踪自适应波形选择算法。在雷达成像方面,研究了以降低成像处理复杂度为目的的自适应发射分集技术和基于发射分集的曲线弹道合成孔径雷达(Synthetic Aperture Radar,SAR)成像算法。首先,针对以景象匹配制导为目的的曲线弹道SAR成像问题,提出通过自适应调整雷达波束指向使雷达近似工作于等效正侧视几何下,能够大大减小距离徙动校正难度及回波信号方位向和距离向耦合,并在此基础上设计了利用部分孔径数据和高效频谱分析方法的RD-Dechirp(Range Doppler-Dechirp)快视成像算法;其次,针对以寻的制导为目的大斜视曲线弹道SAR成像问题,通过在雷达发射信号中引入自适应调整的非线性调频分量,提出了曲线弹道SAR非线性调频变标(Nonlinear Chirp Scaling,NCS)成像算法,能够同时处理大斜视曲线弹道情况下距离徙动校正和二次距离压缩的方位频率依赖性和距离空变性,算法计算量与标准CS算法及RD算法相当,提供了成像性能与计算量的完美折中;最后,针对任意航迹和斜视几何情况下的聚束SAR成像问题,提出了基于PRI(Pulse Repetition Interval)自适应调整的PFA(Polar Format Algorithm)成像算法,能够避免原始PFA算法的插值操作和利用全部波数域数据进行成像,生成高质量SAR图像仅需要FFT和复乘运算,能够兼顾平台航迹灵活性、成像精度及成像效率。论文充分利用雷达接收系统到发射系统的反馈机制和先验知识,研究了面向目标参数估计问题的雷达发射系统自适应调整技术和方法,研究成果有助于从根本上改变现有雷达系统的工作方式,提高现有雷达系统的工作性能和复杂环境适应能力,为推动我国雷达系统的智能化和认知雷达技术的实用化提供基础理论和关键技术支撑。
[Abstract]:The transmission waveform and working mode of the existing radar system are generally fixed, and the receiving system only passively handles radar signals and data, so that even carefully designed radar systems can not obtain ideal processing performance in complex spatio-temporal environment. The cognitive radar has closed loop feedback mechanism to the transmitter to the transmitter. According to the prior and real-time environment and target knowledge, the adaptive dynamic adjustment of radar transceiver system has fundamentally changed the one-way information processing mode of the traditional radar, and can greatly improve the performance of the radar system and the adaptability of the environment. It has important theoretical research and military application value. It has been widely used by many experts and scholars at home and abroad and the military. The adaptive transmission waveform optimization technique is one of the key technologies of the cognitive radar system. The paper mainly studies the adaptive optimization and adjustment technology of radar launching system for target parameter estimation. The specific content can be divided into target tracking (moving target state estimator) and radar imaging (the static target electromagnetic scattering coefficient). The two aspects: in target tracking, the adaptive waveform selection technique for improving the tracking precision of maneuvering target is studied in the environment of clutter. Firstly, the process of radar receiving system and tracking measurement data acquisition is modeled in detail, and the data statistics of radar emission waveform for different signal to noise ratio are analyzed. The influence of the characteristic and the quantitative calculation method, and design the CF-HFM (Constant Frequency-Hyperbolic Frequency Modulation) radar pulse string similar to the bats' acoustic signal in the natural world, and put forward the adaptive waveform selection algorithm for the maneuvering target tracking based on the radar task and the adaptive emission waveform and detection. In order to accurately estimate the posterior probability density function of the mixed motion state of the maneuvering target, an interactive multi model probabilistic data association particle filter (Interacting Multiple Model Probabilistic Data Association Par) is proposed for the precise estimation of the post probability density function of the mixed motion state of the maneuvering target. Ticle Filter, IMMPDAPF), it is estimated that the cycle only needs the basic state particles, can control the number of particles corresponding to each motion model, and introduces the good robust Amari alpha divergence to characterize the information contained in the radar measurement data of different emission waveforms, and puts forward the information theory based on the information theory based on the clutter environment which is closely combined with the IMMPDAPF implementation process. The adaptive waveform selection algorithm for maneuvering target tracking is proposed. In the aspect of radar imaging, the adaptive transmit diversity technique which aims to reduce the complexity of imaging processing and the Synthetic Aperture Radar (SAR) imaging algorithm based on the emission diversity are studied. In line ballistic SAR imaging, it is proposed that using adaptive adjustment of radar beam direction to make radar approximate to the equivalent positive side view geometry, it can greatly reduce the difficulty of distance migration correction and the azimuth and range coupling of echo signals. On this basis, a RD-Dechirp (Ran) using partial aperture data and high efficiency spectrum analysis method is designed. Ge Doppler-Dechirp) fast vision imaging algorithm; secondly, aiming at the problem of SAR imaging of large squint trajectory with the aim of homing guidance, a nonlinear frequency modulation (Nonlinear Chirp Scaling, NCS) imaging algorithm of curve ballistic SAR (Nonlinear Chirp Scaling, NCS) is proposed by introducing the adaptive adjusting nonlinear frequency modulation component in the radar transmitting signal. Considering the azimuth frequency dependence and distance space variability of distance migration correction and two distance compression in the case of curve trajectory, the computational complexity of the algorithm is equivalent to the standard CS algorithm and the RD algorithm, which provides a perfect compromise between the imaging performance and the computational complexity. Finally, the PRI (P) is proposed for the cluster SAR imaging problem under arbitrary track and squint geometry. Ulse Repetition Interval) adaptive PFA (Polar Format Algorithm) imaging algorithm can avoid the interpolation operation of the original PFA algorithm and use all the data of the wavenumber domain to imaging. The generation of high quality SAR images requires only FFT and multiplicative operation. It can give consideration to the flexibility of the platform track, the imaging precision and the imaging efficiency. The paper is fully profitable. By using the feedback mechanism and prior knowledge of radar receiving system to the launch system, the adaptive adjustment technology and method of radar launching system for target parameter estimation are studied. The research results help to fundamentally change the working mode of the existing radar system and improve the performance of the existing radar system and the adaptability of the complex environment. It provides basic theoretical and key technical support for promoting the intellectualization of radar system and the practicality of cognitive radar technology in China.

【学位授予单位】：国防科学技术大学
【学位级别】：博士
【学位授予年份】：2014
【分类号】：TN958

【相似文献】