【作者】 陆必应；

【导师】 梁甸农；

【作者基本信息】 国防科学技术大学 ， 信息与通信工程， 2006， 博士

【摘要】 天基雷达实现地面运动目标指示功能具有重要的意义,其信号处理方法已成为目前雷达界研究的热点。本文以天基多通道、多基地GMTI雷达为应用背景,系统地研究了天基稀疏孔径空时自适应处理面临的基本问题和距离、多普勒、角度模糊对GMTI雷达性能的影响与消除方法。全文的主要工作如下:第二章研究了天基稀疏阵空时自适应处理的基础性问题。在分析分布式小卫星编队构形和单星雷达超大孔径天线结构的基础上建立了天基GMTI雷达天线阵列模型:稀疏子阵列和稀疏子天线,随后基于空时等效理论分析得到了预测两种阵列杂波自由度的简单准确方法,在此基础上分析了天基雷达实现GMTI的基本约束条件。研究了适用于天基雷达的高效STAP算法结构与处理流程,从理论上推导了受地球自转影响的天基雷达杂波多普勒表达式,研究了地球自转对天基稀疏阵STAP性能的影响与消除方法,结果表明简单的距离向回波多普勒校正即可有效消除其影响。第三章从天基雷达回波多普勒历程出发分析了距离、多普勒角度模糊对天基稀疏阵STAP和GMTI性能的影响,纠正了已有结论的片面性和局限性,指出距离模糊阻碍距离向回波多普勒校正,恶化最小可检测速度,引起输出SCNR损失;多普勒角度模糊对GMTI性能的影响与杂波先验信息和阵列结构紧密相关。第四章研究了基于阵列信号处理理论的角度模糊抑制方法。改进了迭代线性约束最小二乘方向图综合方法,使其适用于随机稀疏阵;通过理论推导将随机稀疏阵的方向图综合问题转化为二次锥规划问题,利用高效的内点法求解。基于上述方法研究了方向图特性与阵列结构的关系,为消除栅瓣测速盲区建立了基础。提出了基于阵列信号处理理论的角度模糊抑制方法:对均匀稀疏阵阵元位置施加随机扰动消除测速盲区,利用波束陷零法抑制虚假目标。给出了两种空时二维处理器的波束陷零方法——线性约束自适应空时处理器和人工干扰波束陷零法以,提出了一种重影空时导向矢量估计方法,仿真结果表明模糊抑制方法稳健、有效。第五章研究了基于可变脉冲重复周期的距离、多普勒模糊抑制方法。对多重PRF组,针对多目标应用背景,提出采用参量加权的聚类分析方法区分真实目标与重影,利用简单的遗传算法进行PRF组的优化设计,获得了良好的消除测速测距盲区、抑制重影和解模糊的效果。对参差脉冲重复间隔(PRI),提出了两种改变PRI的方式——固定PRI加随机扰动和均匀递增扰动,仿真分析表明二者均能有效抑制测速盲区,同时引起约2dB的输出SCNR损失。第六章研究了频率正交波形解多普勒、角度模糊的方法。从理论上分析了解模糊对正交信号个数和频差等波形参数的约束,研究了相关参数对空时频自适应处理器输出SCNR的影响。提出了综合利用频率正交信号空时频自适应处理和不同频率信号分别空时自适应处理结果的二级检测解模糊方法,仿真结果表明该方法能有效解模糊,同时降低虚警概率和漏警概率。更多还原

【Abstract】 Ground Moving Target Indicator (GMTI) using Space Based Radar (SBR) is attractive and the signal processing theories are studied by many researchers around the world. With the space based multi-channel and multi-static radar as the background, this dissertation focuses on the fundamentals of space based Space-Time Adaptive Processing (STAP), the effects of range, doppler and angle ambiguities on ground moving target indicator radar and the mitigation methods. The main achievements of this paper are as follows:Elementary problems for space based STAP are studied in chapter 2. Based on the analysis of formation configurations of distributed small satellites system and the very large aperture antenna geometries of single satellite based radar system, the antenna arrays of space based GMTI radar are modeled as sparse subarrays and sparse array with non-isotropic elements. Simple and accurate clutter rank estimating formulas are derived for these two sparse array models using space time equivalence theory. Basic constraints of space based GMTI radar are presented based on the formulas. Then efficient STAP algorithms and architectures are provided according to the limitations of SBR. The earth rotation effects on STAP performance are analyzed with a formula for clutter Doppler returns affected by the earth’s rotation given. And a simple mitigation method based on range-doppler compensation is provided.Special effects of range, Doppler and angle ambiguities on the performance of space based STAP and GMTI are analyzed based on echo Doppler in chapter 3. Original conlusions are developed forward and described as follows: Range ambiguity obstructs range-doppler compensation of clutter returns, degrading the Minimum Detectable Velocity (MDV) and output Signal to Clutter plus Noise Ratio (SCNR) and increasing false alarm probability. Apriori clutter information and array geometry fluctuate the Doppler and angle ambiguities effects on GMTI performance.Angle ambiguity mitigation methods based on array signal processing theory are studied in chapter 4. Two random sparse array pattern synthesis approaches are lucubrated. Firstly the iterative linear constraints least square method are revised and made applicable to random sparse array. Then the problem is expressed as second order cone programming which is solved with interior point method. The approaches are used to analyse the influence of array geometries on array patterns, which gives a direction to suppress velocity bind zone induced by grating lobes. The angle ambiguity mitigation method is described as follows: The element positions of sparse uniform array are randomized to reduce blind speed zones, then the so called "nulling STAP" algorithm is used to suppress false targets. The linear constraints STAP method and the man-made interference method are provided to form nulls in the directions of ghosts. Experimental results illustrate that the ambiguity mitigation method is robust and effective.Range and Doppler ambiguity suppression methods using variable Pulse Repetition Frequency (PRF) are studied in chapter 5. When multiple PRF sets are used, a weighted clustering algorithm is provided to distinguish the real targets from ghosts if multiple targets detected. Simple evolutionary algorithms are used to select appropriate PRF sets. The range and velocity bind zones, ambiguities and the ghosts are eliminated effectively using this method. Two Pulse Repetition Interval (PRI) variation schemes, the random disturbance and the uniform step disturbance, are added to the constant PRI when staggered PRI waveform is used. Simulation results show that preferable velocity coverage is achieved at the price of about 2dB output SCNR degradation for both the schemes.Doppler and angle ambiguity mitigation methods using frequency orthogonal waveform are studied in chapter 6. Limitations of waveform parameters, such as frequency gap and the number of signals, are analyzed and their effects on the output SCNR of space time frequency adaptive processor are addressed. T hen a two-step detection algorithm is presented to reject Doppler and angle ambiguities, which collaborates the results of the space time frequency adaptive processor and space time adaptive processors of each frequency. Simulation results illustrate that the algorithm is effective in resolving Doppler and angle ambiguities, decreasing both the false alarm probability and the leak alarm probability.更多还原