【作者】 李会勇；

【导师】 何子述；

【作者基本信息】 电子科技大学 ， 信号与信息处理， 2009， 博士

【摘要】 多功能相控阵雷达,即用一部雷达系统可实现多部雷达的战术功能,是相控阵雷达的发展方向。为了使多功能相控阵雷达具有目标识别、成像的能力,这就要求相控阵雷达应具有高的距离分辨率,即相控阵雷达应工作在宽带。宽带数字阵雷达为实现相控阵雷达的多功能提供了可行的硬件平台。它用数字时延取代了传统的移相器,可减小孔径效应,使相控阵雷达实现宽带宽角扫描。另外,它还具有很高的动态范围、容易实现发射波形和频率捷变等优点。宽带数字波束形成是宽带数字阵雷达中的关键技术之一。要实现宽带波束形成,不能再采用普通相控阵雷达中的移相技术,而必须采用数字时延,另外,为了在形成正确波束指向的同时,还在干扰到来方向形成零点,以提高系统性能,所采用的自适应算法和技术也不同于窄带自适应波束形成技术。所以,以宽带数字阵雷达为应用背景,开展数字时延产生、宽带波束形成、自适应宽带干扰置零、系统实现方案等关键技术研究,有着重要的理论意义和实用价值,这也是目前阵列信号处理研究中的热点问题。本文针对宽带数字阵雷达系统波束形成技术进行了研究,主要研究包括:1、研究了基于分数时延的宽带波束形成技术,分析了直接射频采样模式下,基于时延的波束形成结构,并对分数时延的三种产生方法在宽带波束形成中的应用与实现、性能及复杂性进行了分析、比较,指出尽管分数时延能够获得较理想的阵列方向图特性,但是全部阵元都采用分数时延结构可能会带来系统实现的复杂性和高成本。2、研究了分子阵的宽带波束形成技术。在实际大型相控阵雷达中,由于阵元数成百上千甚至上万个,如果每个阵元后面都采用分数时延方法,会给系统海量数据的传输与存储、高速滤波器实现等方面带来很大的压力,必须采用分子阵的波束形成结构以降低系统实现难度和成本。研究了子阵划分的原则和方法,对基于阵元内单元移相加子阵间分数时延的宽带波束形成性能进行了分析,并用仿真结果验证了方法的有效性。3、研究了宽带自适应波束形成算法在数字阵中的应用。宽带自适应波束形成方法不同于窄带波束形成,论文对宽带自适应波束形成算法,特别是ISM和CSM算法在宽带数字阵中的自适应干扰置零方面的性能进行了分析、对比,对它们在不同工作频率、工作带宽、干扰模式、阵列形式等条件下的性能进行了仿真分析,并讨论了快拍数、采样率、FFT点数等对干扰置零性能的影响,这些结果和结论为宽带数字阵实验系统选择自适应波束形成算法以及系统参数提供理论依据。4、研究了宽带数字阵雷达可能的实现方案及对应的波束形成方法。分析比较了宽带数字阵雷达可能采用的三种宽带信号:瞬时宽(频)带信号、宽带线性调频信号和频率步进信号,以及直接射频采样、中频采样等方式下宽带数字阵雷达系统及波束形成实现方案及特点、难度和存在的问题。对于去斜方案,提出了一种能够灵活控制波束指向的宽带波束形成方法。另外,针对宽带数字阵实现时可能存在的多通道采样不同步问题,分析了它对波束形成性能的影响,并进行了详细的理论推导和分析,以及仿真验证,为今后宽带数字阵雷达系统设计提供硬件选型和波束参数选择等方面的理论参考。此外,论文还针对阵列处理的另一个重要应用领域—MIMO通信系统中的检测算法进行了研究。MIMO技术是近年来的一个热点研究问题,也是下一代无线通信系统(B3G)中的关键技术之一。论文研究了非线性检测算法在MIMO通信系统中的应用。利用非线性算法将MIMO系统中多个天线接收信号进行合成,充分利用接收信号中所包含的高阶信息,提高MIMO系统检测性能。为了克服非线性检测算法在计算复杂性方面的不足,提出了两种降低计算量的方法:基于矩阵递增求解的方法和基于Fisher比的稀疏化方法。仿真验证了这两种算法都能够显著降低非线性算法的计算量,而且误码率性能远远超过传统的常规MIMO系统线性检测算法。更多还原

【Abstract】 Multifunction digital array radar, which employs a single radar system to obtain the tactic functions of multiple radars, is very important for the development of future phase array radar. High range resolution is required to enable the multifunction digital array radar for target identification and target imaging. Thus, the operation frequency of the phased array radar should be wideband.Wideband digital array radar makes the realization of the multifunction phased array radar possible. In the wideband digital array radar, time delay elements are used instead of the phase shifters used in the traditional phased array radars, such that the aperture effect can be eliminated and wideband wide scan can be implemented. Further, the wideband digital array radar has the advantages of very high dynamic range, facilitating transmitting waveforms, and frequency agility.Wideband digital beamforming is the key technique for the wideband digital array radar. Wideband beamforming requires replacing phase shifters used in the traditional phased array radars by the time delay elements. On the other hand, to form a null in the direction of interference to improve the system performance, adaptive beamforming algorithms and techniques are needed, which are different from those used for narrowband adaptive beamforming.Towards the application of wideband digital array radar, the research on digital time delay, adaptive wideband interference nulling, and other related techniques are of significant theoretical and practical importance. Of course, they are emergent topics in the current research of array signal processing.This thesis studies the beamforming for the wideband digital array radar. The research includes:1. Wideband digital beamforming based on fractional time delay is investigated. When direct RF sampling is adopted, the beamforming scheme based on time delay is discussed. Three types of fractional time delay approaches are applied to the wideband beamforming. The performance and complexity of these approaches are analyzed and compares. The complexity of applying fractional time delay to all array elements is shown.2. Beamforming based on sub-array level is discussed. Although fractional time delay can achieve accurate digital time delay and obtain the desired array beam pattern, when the number of array elements is very large, as in some practical phased array radars, employing fractional time delay for each array element seems impossible due to the problems faced in the data transmission and storage and high speed filtering. In this case, we have to resort to the beamforming scheme in the sub-array level. The principle and approach for dividing subarrays are studied. The performance of wideband beamforming based on applying phase shifters within sub-array and fractional time delayer between sub-arrays is analyzed and illustrated by simulations.3. Wideband adaptive beamforming algorithms are studied. The performances of the wideband adaptive beamforming algorithms, particularly the ISM and CSM algorithms, used in the wideband digital array radar system, are analyzed, simulated, and compared. The performance of these algorithms under different operation frequencies, operation bandwidths, coherent interferences, and array configurations are investigated through simulations. The discussions on the impacts of the number of snapshots, sampling frequency, and FFT point on the beamforming performance are provided, which is useful for the test system design.4. The possible realizations of the wideband digital array radar and the corresponding beamforming methods are investigated. Three types of wideband signals are analyzed and compared, including instantaneous wideband signals, wideband linearly modulated signals, and frequency stepped signals. The advantages, difficulties and existing problems of the realization of wideband digital array radar system based on the direct RF sampling and IF sampling are studied. A wideband beamforming approach based on dechirping is proposed. Further, the impact of the multi-channel non-synchronized sampling on the beamforming performance is studied, which provides a theoretical reference for the hardware and waveform selection for designing the wideband digital array radar system. Additionally, the application of the array signal processing in MIMO communications system is studied in this thesis. MIMO technique is a hot topic in recent years, which is one of the key techniques of the next generation wireless communications systems (Beyond 3G).We focus on the application of the nonlinear detection algorithms in MIMO communications system. Multiple received signals in the MIMO system are combined using nonlinear algorithms. Taking the advantage of the higher-order information, the detection performance of the MIMO system is improved. To combat the drawbacks of the nonlinear algorithms in computational complexity, two methods based on incremental method and Fisher rate method are proposed for reducing the computation load of the nonlinear algorithms, which also lead to improved bit error rate performances than the traditional linear detection algorithms used in MIMO systems.更多还原