MIMO雷达信号处理技术及实现的研究

发布时间：2018-05-14 06:18

  本文选题：MIMO雷达 + 长时间积累　； 参考：《北京理工大学》2015年博士论文

【摘要】：多输入多输出雷达(MIMO radar,Multiple-Input Multiple-Output radar)利用多个发射天线发射信号和多个接收天线接收目标回波信号进行处理,并通过分集增益获得了一些优良特性。这些优点使其获得了良好的应用前景。本文从以下几个方面进行了研究。其一,研究了MIMO雷达的长时间积累算法。MIMO雷达进行长时间积累的优势在于它可以进行几百以至于上千个积累脉冲数的积累而不需要考虑跨波束的问题。由于它是通过计算来形成发射波束和接收波束,发射波束为多个宽波束,甚至可以充满整个空间以凝视方式工作,而无需波束扫描。在非相参积累方面,主要针对Hough变换进行了改进,提出了快速方法和机动运动目标积累检测方法;在相参积累方面对于包络插值移位补偿和频域联合补偿方法进行了改进,并提出了采用Zoom-FRFT等分级FRFT进行速度和加速度局部细估的方法。在MIMO雷达实现中,综合空间积累及本文中所提的慢时间积累方法来实现多目标情况下的机动弱目标的搜索和跟踪的性能优于相控阵体制雷达。其二,研究了MIMO雷达基于合成宽带的高分辨成像技术。MIMO雷达进行合成宽带高分辨成像的优势在于可同时接收多个子带信号合成宽带从而提高了成像效率和效果。本文提出了改进的频域合成方法,这种方法不仅适用于小场景也适用于大场景的应用。本文还在参数选择、栅瓣和旁瓣抑制、幅度和相位校正及运动补偿等几个方面进行了分析,提出了适合于工程应用的实现途径。在高分辨检测方面本文结合文献中提出对于低信噪比下基于神经网络的检测方法。相比已有方法本文中的方法通用性强且适用于工程实现。其三,研究了基于VPX总线及TMS320C6678多核DSP硬件平台的MIMO雷达信号处理软件的高性能实现问题。相比基于CPCI总线和单核DSP平台的系统实现,本文中的系统实现在并行性能、运算能力和成本等方面都具有明显优势。但是,由于新平台自身的复杂性,且缺乏成功的借鉴经验,因此在初始开发阶段存在很多难点需要攻克。本文在介绍信号处理系统设计、硬件平台和支撑软件设计的基础上对MIMO雷达信号处理软件设计方法和关键问题进行了深入而细致的研究并给出了实现方案和解决方法,并在项目组搭建的MIMO雷达原理实验系统的基础上进行了实验和验证。文中对实验目的、实验条件、实验原理进行了介绍,并给出了实验结果。实验结果表明了MIMO体制雷达的优势及研究算法的有效性和实用性,同时也证明了MIMO雷达信号处理软件并行设计的可行性和高效性。
[Abstract]:Multiple-Input Multiple-Output radar is processed by transmitting signals from multiple transmit antennas and receiving echo signals from multiple receiving antennas, and some excellent characteristics are obtained by diversity gain. These advantages make it have a good application prospect. This article has carried on the research from the following several aspects. Firstly, the advantage of long time integration algorithm of MIMO radar is that it can accumulate hundreds or even thousands of integrated pulses without considering the problem of cross beam. Because it is calculated to form transmitting and receiving beams, the transmitting beams are multiple broad beams, which can even fill the entire space to work in a staring manner without beam scanning. In the aspect of non-coherent integration, the paper mainly improves the Hough transform, proposes a fast method and a mobile target accumulation detection method, and improves the envelope interpolation shift compensation and frequency domain joint compensation method in the aspect of coherent integration. A method of local fine estimation of velocity and acceleration using Zoom-FRFT and FRFT is presented. In the implementation of MIMO radar, the performance of the search and tracking of maneuvering weak targets with multiple targets is superior to that of phased array radar by integrating the space accumulation and the slow time accumulation method proposed in this paper. Secondly, the advantages of synthetic wideband high resolution imaging technology based on synthetic wideband for MIMO radar are that it can receive multiple sub-band signals to synthesize broadband at the same time, thus improving the imaging efficiency and effect. In this paper, an improved frequency domain synthesis method is proposed, which is suitable not only for small scenes but also for large scene applications. In this paper, parameters selection, gate lobe and sidelobe suppression, amplitude and phase correction and motion compensation are also analyzed, and a suitable approach for engineering applications is proposed. In the aspect of high resolution detection, a neural network-based detection method for low signal-to-noise ratio (SNR) is proposed. Compared with the existing methods, the method in this paper is universal and suitable for engineering implementation. Thirdly, the high performance realization of MIMO radar signal processing software based on VPX bus and TMS320C6678 multi-core DSP hardware platform is studied. Compared with the system implementation based on CPCI bus and single core DSP platform, the system implementation in this paper has obvious advantages in parallel performance, computing power and cost. However, due to the complexity of the new platform and the lack of successful experience, there are many difficulties in the initial development phase. On the basis of introducing the design of signal processing system, hardware platform and supporting software, the design method and key problems of MIMO radar signal processing software are studied deeply and meticulously in this paper, and the realization scheme and solution are given. Based on the experimental system of MIMO radar principle built by the project team, the experiment and verification are carried out. The purpose, conditions and principle of the experiment are introduced, and the experimental results are given. The experimental results show the advantages of the MIMO radar and the effectiveness and practicability of the algorithm. At the same time, the feasibility and efficiency of the parallel design of the MIMO radar signal processing software are also proved.
【学位授予单位】：北京理工大学
【学位级别】：博士
【学位授予年份】：2015
【分类号】：TN957.51
，

本文编号：1886682