基于杂波结构的MIMO雷达精细化检测方法及波形设计

Radar target detection in heterogeneous and strong clutter has wide applications in the fields of public and national security. Conventional adaptive detection of radar targets employs homogeneous training data to estimate clutter covariance matrix for suppressing clutter. In practice, training data are insufficient and heterogeneous, which obviously deteriorates the performance of conventional adaptive detection algorithms, and even makes them unable to work. This project focuses on the problem of detecting radar targets in heterogeneous and strong clutter. Adaptive detection algorithms are proposed by exploiting structures of clutter covariance matrix in collocated multiple-output multiple-input (MIMO) radar, without resorting to training data. In addition, transmit waveforms are optimized by incorporating structures of clutter covariance matrix, according to the criterion of maximizing detection probability. The detection performance can be greatly improved. This project will build a semi-physical simulation verification platform, and carry out simulations and feasibility tests of MIMO radar detection and waveform design. This project fully exploits the potentials of clutter structures, and realizes adaptive detection without training data, which avoids serious problems due to the heterogeneity and insufficiency of training data. Therefore, the project has important scientific significance and application value.

非均匀强杂波背景下的雷达目标检测在公共和国防安全领域应用广泛。传统雷达目标自适应检测算法利用均匀训练数据估计杂波协方差矩阵来抑制杂波。实际中，训练数据的匮乏和异质将会导致传统自适应检测算法的性能明显下降，甚至无法工作。本项目针对非均匀强杂波环境下的雷达目标检测问题，在集中式MIMO雷达中利用杂波协方差矩阵结构设计不需要训练数据的自适应检测算法，基于检测概率最大化准则融合杂波协方差矩阵结构信息优化发射波形，显著提升检测性能。搭建半实物仿真验证平台，完成MIMO雷达检测和波形设计性能的仿真及可行性试验验证。本项目充分挖掘杂波结构信息的应用潜力，实现不需要训练数据的自适应检测，克服由训练数据匮乏和异质带来的问题。因此，本项目具有重要的科学意义和应用价值。

传统雷达目标自适应检测算法利用均匀训练数据估计杂波协方差矩阵来抑制杂波。实际中，训练数据的匮乏和异质将会导致传统自适应检测算法的性能明显下降，甚至无法工作。本项目针对非均匀强杂波环境下的雷达目标检测问题，在集中式MIMO雷达中利用杂波协方差矩阵结构设计了多种不需要辅助数据的自适应检测算法，显著提升了雷达目标检测性能。共发表SCI期刊论文57篇（其中IEEE Trans论文26篇），会议论文6篇，撰写英文学术专著2部，授权发明专利1项。该项目的研究提高了我国复杂环境下雷达目标检测的基础研究水平和自主创新能力，可应用于MIMO雷达系统中，具有重要的科学意义和应用价值。