极化雷达波形设计及STAP算法研究

In order to further improve the diversity advantage of MIMO radar, improve the performance of clutter interference suppression when the angle between clutter interference and desired signal is small, enhance the performance of radar target detection, the polarization characteristics is introduced into the design of MIMO radar system. By making full use of the advantages of different characteristics between desired signal and clutter interferences in the space, time, frequency and polarization domain, the model of the received polarization-space-time echo signal, the methods of the optimum design of the polarization waveform and the adaptive processing of the polarization-space-time domain are studied. Based on the mutual information theory and convex optimization theory, a series of fast solving algorithms, such as alternating projection optimization method of multi dimension space and maximal-minimal algorithm, are used to realize the design of polarization waveform. By using the low rank and block Toeplitz structure of the space-time clutter covariance matrix, the covariance matrix reconstruction is realized by using the compressed sensing technology and the non negative matrix factorization technique, which reduces the complexity of the polarization-space-time adaptive processing algorithms, and improves the practicability of the algorithms. Make full use of polarization characteristics at the same time, to further explore the high efficiency and low complex degree of waveform design methods and clutter interference suppression methods, to form a set of relatively complete theory and methods for polarization radar, to provide the theory support for the polarization radar waveform design and clutter interference suppression.

为进一步改善普通MIMO雷达的分集优势，提高杂波干扰与期望信号夹角较小情况下的杂波干扰抑制能力，增强雷达的目标检测性能，将极化特性引入MIMO雷达系统设计中。利用有用信号与杂波干扰信号在空域、时频域以及极化域存在的差异，研究接收极化-空-时回波信号的建模，极化雷达波形优化设计的方法以及极化-空-时域联合的自适应处理等内容。利用互信息理论和凸优化理论，采用多维空间交替投影优化方法以及极大-极小算法等一系列快速求解算法来实现极化波形和基带发射波形设计；利用空时杂波协方差矩阵的低秩、块Toeplitz结构，采用压缩感知技术和非负矩阵分解技术实现协方差矩阵重构，实现极化空时自适应处理算法的降维降秩。在充分利用极化特性的同时，提出一系列高效低复杂度的波形设计和杂波干扰抑制新方法和新算法，为极化MIMO雷达的实用提供前期理论支撑和储备。

多域联合处理的抗干扰算法能够在不增加阵列大小的基础上增加阵列抗干扰自由度，充分利用所联合的极化域/频域/空域信息有效区分抑制干扰。本课题对于提升多域联合处理干扰抑制算法的高动态稳健性和可实现性做了如下工作：.首先，给出空时/空时极化多域联合处理信号模型，并将斜投影滤波、线性约束最小方差（LCMV）和功率倒置（PI）算法推广到多域联合处理；.其次，考虑到多域联合处理抗干扰算法在高动态下可能面临零陷失配的问题，针对空时处理进程中，干扰零陷失配扰动角度服从期望为0的高斯分布的情形，提出了无需干扰先验信息、计算简单、零陷展宽宽度可调的空时协方差矩阵锥ST-CMT；.然后，针对多域联合处理的处理数据维度增大的问题，将联合迭代优化（JIO）降秩方法与多域联合干扰抑制算法相结合，降低了数据处理的维度并不再需要矩阵求逆，大大增加了多域联合抗干扰算法的可实现性；通过对输出功率与预设门限比较，可选择性更新降秩变换矩阵和降秩后权向量，进一步降低了联合迭代优化降秩抗干扰算法的更新次数，并自适应调整步长；.另外，也综合考虑联合迭代优化降秩和零陷展宽。将协方差矩阵锥分别与联合迭代优化降秩PI-JIO和变步长联合迭代优化降秩PI-JIOV相结合成：PI-JIO-CMT和PI-JIOV-CMT，可在降低多域联合干扰抑制算法计算复杂度的同时进行零陷展宽；.最后，基于杂波空间时间谱的稀疏性，研究了将稀疏恢复技术应用于STAP中，与传统的降维STAP和降秩STAP相比，SR STAP需要较少的训练样本就能达到同样的性能，减小了非均匀杂波对杂波抑制性能的影响。