摘 要 摘 要
高频地波雷达是一种可以有效探测海面超视距隐身目标的雷达系统，通常采用宽波束照射、窄波束单方向接收目标回波。由于目标特性在高频段处于瑞利区前段或者谐振区，不确定的目标姿态及RCS起伏极易导致回波微弱而丢失目标。为了克服目标的闪烁，本研究充分利用目标不同频率和角度RCS的差异，将多输入多输出（MIMO）体制与高频雷达相结合，利用MIMO体制的优势来提高雷达系统时间、空间、频率管理的自由度，有效综合多方向、多载频的目标回波信息，达到改善系统分辨力的目的，进而提高雷达系统的探测能力。因此，关于MIMO体制与高频地波雷达结合的关键技术研究具有重要意义。
本研究将MIMO的体制结合高频雷达的自身特点，建立了区别一般MIMO体制的多载频MIMO高频雷达模型。本研究分别建立了集中式和分布式MIMO雷达模型，针对不同模型提出了有效的信号处理算法，得到了一些具有积极意义和参考价值的方法和结论。总体来说，本研究主要包括以下几个方面：
第一，研究并改进了MIMO模糊函数使其适应评估复杂的MIMO雷达系统速度距离分辨能力，突破了传统模糊函数理论的局限性。本文提出非连续谱MIMO高频雷达模糊函数：通过研究集中MIMO模型分集发射不同载频信号，对系统空域与时域耦合问题进行了研究，有效评估该系统角度距离分辨力，得到了非连续谱波形参数选择准则；进一步在多载频分布式MIMO模型下，通过扩展传统模糊函数的研究评估系统空间分辨力，得到了分布式MIMO雷达的传感器布放一般规律。本研究提出的改进MIMO模糊函数对于高频MIMO雷达全局分辨力分析具有重要的意义，为后续关键问题的提出及研究打下了坚实的基础。
第二，为了有效提高高频雷达距离分辨力，针对非连续谱信号产生的较高距离旁瓣的缺点，本研究提出了两种非连续谱信号来解决较高旁瓣问题——接收端非连续谱旁瓣抑制和发射端波形设计。区别于以往非连续谱波形时分复用构造形式，本研究采用天线分集发射不同载频相位编码信号，在接收端将非连续谱信号相参拼接，对相参处理后主瓣附近产生的较高旁瓣进行抑制。基于凸优化算法设计旁瓣抑制滤波器，对有限长数据产生的伪峰滑窗剔除，有效的解决了多目标环境下弱目标被较高副瓣遮挡问题。本算法实时性好且简单实用，便于工程实现。基于发射端波形设计，本研究利用波形平均通带阻带功率比与自相关旁瓣关系，提出波形优化的代价函数。通过量子遗传算法优化低自相关旁瓣波形，仿真实验取得了预期的效果，优化的波形实现了频谱约束下最优副瓣等级。本章关于非连
- I-
哈尔滨工业大学工学博士学位论文 续谱信号的研究对于提高高频雷达恶劣电磁环境工作问题具有重要意义。
第三，针对高频雷达的角度分辨力差问题，区别于以往MIMO模型虚拟阵列技术，提出通过多载频聚焦算法虚拟大的阵列孔径。针对虚拟孔径的单元间距大于半波长，阵列综合产生栅瓣的问题，本研究进一步提出了基于预条件共轭梯度法重构非均匀空域信息，有效抑制因阵列稀疏产生的栅瓣问题。在重构标准频率下均匀空域信息时，需要求解高维方程，预条件的引入显著提高了算法的收敛速度。同时，针对不同频率下，目标不一致的幅度相位响应、回波信噪比及多普勒信息，提出了相应的预处理流程。此外，分析了发射载频选择对空域综合性能的影响。多载频MIMO空域综合提出了一种新的信号融合思路，有效扩展了高频雷达的孔径尺寸。
第四，本研究建立了多载频分布式MIMO高频雷达模型，针对该模型产生空间和多普勒域稀疏采样问题，提出一种基于压缩感知分布式雷达系统算法。本研究采用非均匀周期发射多载频相位编码信号，有效减少多脉冲雷达系统不同节点数据传输负担。针对稀疏回波信号恢复问题，构造统一的空域和多普勒域感知矩阵，有效解决了分布式系统多站点回波信息集中处理问题，及分布式系统产生的目标定位模糊。针对空间网格细化导致压缩矩阵相关性变强，将多载频信息的引入，较好的解决了感知矩阵空间等距条件的限制。此外，针对实际情况中，不同方向产生的未知相位响应无法准确重构问题，提出测量矩阵附加目标未知的相位响应来改善目标定位误差。最后，讨论了感知矩阵的列向量相关性，系统估计误差性能与脉冲个数的关系，以及空间分辨网格的划分准则。本研究对于感知压缩算法应用于实际的分布式MIMO雷达模型提供了重要理论依据。
关键词：MIMO雷达；高频地波雷达；非连续谱波形设计处理；多载频空域综合；分布式压缩感知；
- II -
Abstract Abstract
Flight target realizes self-stealth is paradoxical problem for weak target detection. Entering the 21st century, electronic countermeasures and stealth technology has been rapid development. The multiple-input multiple-output (MIMO) systems have been attracted widely attention. Under the conventional microwave detection conditions, Missile target, stealth aircraftthe has weak echo energy, dueing to the characteristics of the structure and size. This category target RCS is obvious divercity under different carrier frequency and illumination angle. Using this feature, MIMO detection system can collect from multiple directions and carrier frequency target echo information, realizing the space diversity, frequency diversity, achieveing the signals energy accumulation and detection. Therefore, MIMO detection technology has a good technical advantage.
In actual radar detection, usually radar working environment is not ideal silent, the electromagnetic environment is fulled of interference. The application of electronic countermeasures also makes electromagnetic environment become more complicated. In this environment, the available signal bandwidth is discontinuous. Especially in the VHF, UHF and band, television, communication signal has become the main source interference. Combined with MIMO technology advantage, the research is about making full use of the discontinuous signal bandwidth, in order to achieving accurate detection purpose. This paper combined MIMO system with the characteristics of high frequency radar, established the multi-carrier frequency MIMO HF radar model. The collocated and distributed MIMO high-frequency radar signal processing is studied, and got some positive meaning method and the conclusion. In general, this paper mainly studied the following problems:
First, the traditional ambiguity function is based on the signal waveform about the target speed and range analysis theory, and for MIMO HF radar system from multi-angle receiving, discontinuous spectrum signal model, the traditional ambiguity function theory cannot accurate analysis of complex multiple-input multiple-output (MIMO) system. This paper introduced target RCS characteristics to the ambiguity function, making the ambiguity function can be analyzed the more generalized space resolution of radar system. This chapter is aimed at discontinuous wave about single static MIMO HF radar system, and analyzing the higher sidelobe resean of the discontinuous spectrum pulse compression; then we analysis distributed MIMO system space resolution through the distributed MIMO ambiguity function and achieving array arrangement law. The MIMO ambiguity is the foundation of the subsequent chaper.
- III -
哈尔滨工业大学工学博士学位论文 Second, in view of the HF band interference is serious, We adopted multiple antenna diversity emission MIMO model. The antenna transmitting discontinuous spectrum cycle phase complementary code signal (MCPC), at the receiving we splited joint discontinuous spectrum phase coded signal, in view of higher sidelobe of the discontinuous spectrum pulse compression near main lobe, we make suppressed processing. Through the convex optimization we design sidelobe suppression filter, and eliminate sliding window pseudo peak due to the limited length data generated, and analysis the waveform property. Based on transmitting terminal optimization waveform, constraint emission waveform spectrum and emission waveform autocorrelation side lobe level level, through the quantum genetic algorithm to optimize the emission waveform. This study provides two kinds of discontinuous spectrum signal used two kinds of modes, and makes the HF radar better adapt to the complicated electromagnetic environment.
Third, the large carrier frequency difference of emission array is adopted to improve the HF radar system spatial resolution, through multiple carrier frequency focusing processing generating virtual larger array aperture. Because of the virtual aperture unit space information data is heterogeneous, drawing lessons from the seismic signal heterogeneous data processing preliminary conditions conjugate gradient method, we reconstructed different carrier frequency domain information. The multi-carrier frequency space information is fused in signal layers. Due to different frequency inconsistent target response of phase and amplitude, the creating virtual array unit receiving data SNR of the also is inconsistent, and different Doppler frequency target registration, we do preprocessing before the spatial synthesized. And the end of the chapter analyzed the emission frequency selection influence spatial synthesized performance.
Fourth, in view of the distributed MIMO high-frequency radar system, it is more difficult to achieve much more the sensor node data signal layer fusion problem, the Compressive sensing algorithm is applied to distributed radar system, effectively solving the problem of the distributed system centralized decision-making, greatly improve the high frequency radar system space resolution, weak target detection ability. Traditional multi-static radar system adopts fusion distributed decision, the strategy each node loss the phase interaction information. The Compressive sensing theory applied to distributed radar system, through the reconstruction all directions echo signal components, to estimate moving target position and speed. According to the different frequency in different directions from the unknown phase response can't accurate reconstruction, we will modify measure the matrix additional unknown target response phase distribution, effectively improve reconstruction error. In the end of chapter
- IV -
Abstract anlysised the system error of estimate performance, and analysis spatial resolution mesh division.
Keywords: MIMO radar, HF surface wave radar, Dsicontinuouse spectrum waveform design and processing, Multi-frequency spatial domain systhesize, Distributed compressive sensing radar
- V -