【作者】 陈辉；

【导师】 黄本雄；

【作者基本信息】 华中科技大学 ， 信息与通信工程， 2009， 博士

【摘要】 阵列天线的高分辨DOA估计技术在雷达、通信、声呐、地震勘探、射电天文等领域都有广泛的应用前景。目前，它已成为无源阵列雷达、智能天线空分多址和声源定位探测等热点领域中的关键技术之一。在其三十多年的发展过程中，国内外学者提出了许多性能优良的方法，取得了丰硕的成果。但是，该领域的研究并不完善，特别是实用性方面的研究还相对落后，仍有很多难题尚待解决。本文着重研究实用化过程亟待解决的相干源问题和阵列校正问题。论文的主要工作概括如下：1．针对相干信号源间的相干系数为1的背景下，提出了一种基于互相关矢量的解相干(CVT)算法和一种基于最大特征矢量的ECVT算法。该类方法相比空间平滑类算法，它不损失阵列孔径，这样它可以保证充分利用阵列孔径来提高相干信号源的DOA估计精度。2．针对CVT算法的局限性，提出了基于广义导向矢量的扩展CVT(GCV)算法，并将其思想进一步推广到广义似然算法(GML)、广义信号子空间拟合(GSSF)和广义噪声子空间拟合(GNSF)算法，这些算法可以实现相干信号源的高精度估计，同时在相干源组数小于阵元数的情况下，算法分辨的信源数大于阵元数。将GCV算法应用到米波阵列雷达上，通过实测数据分析表明GCV算法可以用来进行米波雷达的测高，其估计性能要优于常规空间谱估计解相干算法，如空间平滑算法等。3．针对无源阵列雷达背景，研究了在干扰背景下特殊区间目标的空间信号源分辨问题，提出了利用干扰阻塞和波束空间两种思路来解决强干扰抑制与感兴趣区间目标的检测问题。提出了一种可以抑制固定强干扰的DOA估计方法——干扰阻塞法(IJM)。提出了一种波束空间的解相干算法——BCVT算法，该算法采用解相干与波束形成级联处理方式，其中在波束形成时可以采用自适权或者采用固定深加权的方式来抑制感兴趣区域之外的信号／干扰。4．针对某实际阵列雷达研究了互耦校正问题，提出了一种针对平行线阵的方位估计及互耦自校正算法。该方法可以实现平行线阵在未知互耦参数的情况下，精确且独立估计信号源的方位，然后再估计互耦参数，从而实现了阵列的互耦校正。另外，由于算法的波达方向估计只需一维谱峰搜索，避免了多维联合估计，可明显降低运算量，且算法的估计精度高、互耦校正能力好。最后，通过分析给出了平行线阵解互耦的一些应用条件和特性。5．针对某实际阵列雷达研究了阵列误差的综合校正问题，提出了基于辅助阵元结合辅助信源的综合校正算法，并结合不同误差情况给出了相应的误差校正算法。在阵元位置误差或通道不一致误差存在的情况下，采用ISCP方法可以实现无误差的导向矢量与方位依赖误差的解耦合，从而实现角度与误差的高精度级联估计。在同时存在阵元位置误差和通道不一致误差的情况，可以采用ISCPG方法实现导向矢量与误差的解耦合，从而实现方位估计、阵元位置误差和通道不一致误差的联合估计。分别探讨了多种误差情况下对设置辅助信号源及辅助阵元数的要求。对于米波段的阵列天线，研究了应用ISCG算法来校正阵列天线的通道不一致误差，通过实测数据表明算法可以在未知误差的情况下精确估计信号的角度，从而实现了强干扰背景下的微弱目标检测。更多还原

【Abstract】 High resolution Direction-Of-Arrival (DOA) estimation techniques with arrayantenna have wide applications in a variety of fields ranging from radar,communication,sonar,seismology to radio astronomy.Especially,they become a key technique in thepassive detection of array radar,SDMA of smart antenna system and detection soundsource.Since early 1980s,high resolution DOA estimation techniques have receivedconsiderable attention and a lot of significant progresses have been achieved in this field.However,there are still important and urgent problems that have not been solved perfectlyin their engineering practice.This thesis centers on two problems in practice.One is theDOA estimation of coherent sources and another is array calibration.The maincontributions can be summarized as follows:An algorithm for DOA estimation of coherent source,called correlation vectorToeplitz (CVT) algorithm,is proposed with the correlation coefficient is 1.And anotherECVT algorithm is proposed using maximum eigen-vector based CVT.Compared withspatial smooth algorithms,the new methods have no loss array aperture,and have lowcomputational complexity.Therefore,the new methods have higher estimationperformance.Based on the CVT algorithm,the new Generalized Maximum Likelihood (GML),Generalized Signal Subspace Fitting (GSSF) and Generalized Noise Subspace Fitting(GNSF) are derived.This kind algorithm uses generalized steering vector substitutes forthe conventional steering vector.Therefore,the number of sources resolved by newalgorithm can be larger than the number of sensors as long as the number of the coherentgroup is smaller than the number of sensors.The GCV method was applied to meter radarverify system.The DOA estimation performance was compared using experimental data,demonstrating the effectiveness of new method.Passive radar has attracted great research interest given their importance in amilitary application field.The new interference jamming method (IJM) is presented toestimate the weak target under the condition of presence large power interference of known direction.The method use the information of the strong jamming is known toconstruct a jam matrix,and then estimate signal DOA of the low signal noise ratio in onesector.The method can be applied to the passive array radar based civil broadcastingstation.And the relation of the transform matrix and the restrained interference noisematrix is analyzed.A new orientation estimation and mutual coupling auto-adjust algorithm is proposedto solve the elements mutual coupling problem based on subspace theory and the specialstructure of parallel linear array in this paper.The proposed algorithm can estimateaccurately DOA of signal sources without known the mutual coupling parameters of array.Then the estimated orientation can be used to estimate mutual coupling parameters,whichcan adjust the mutual coupling problem of array.The estimation of DOA only needs onedimension spectrum peak value searching,which avoid the multidimensional jointestimation.The algorithm decreases the computational load,and it has high estimationprecision and good mutual coupling adjusting ability.Based on parallel linear array,a novel Instrumental Sensors Calibration Position(ISCP) method is proposed to estimate the sensor position error or channel gain error.Andanother new method of Instrumental Sensors Calibration Position and Gain (ISCPG) canestimate the sensor position error and channel error same time.The DOA and errorparameter can be completed just using a one-dimensional search or polynomial rooting,with no high-dimensional nonlinear search and convergence burden involved.Another,theISCG method was applied to passive array radar verify system.The proposed algorithmcan estimate accurately target without known the channel gain error of array usingexperiment data were processed to verify systems.更多还原