航空武器数据链阵列抗干扰技术研究

发布时间：2018-03-16 18:24

本文选题：数据链　切入点：阵列抗干扰　出处：《南京航空航天大学》2016年博士论文　论文类型：学位论文

【摘要】：航空武器数据链是用于航空武器目标信息修正、协同控制的专用数据链,是提高航空武器对时敏目标打击能力的关键装备,也是在信息化条件下的体系对抗中取得战术优势的关键一环。航空武器数据链的工作过程是一个逐渐远离己方平台、接近敌方平台的过程,传统抗干扰措施在面对信号捕获速度达到微秒级、干扰等效辐射功率达到兆瓦级,可以实施精确压制的新一代干扰平台时,都将难以取得满意成效。阵列抗干扰技术通过对在空间按一定方式排列的天线阵列所接收的信号进行特定处理,在期望信号方向形成高增益主瓣,增强有用信号能量,在干扰方向形成零陷,抑制干扰信号和噪声,用于航空武器数据链可以大大提高抗干扰能力。由于使用环境的特殊性,阵列抗干扰技术应用于航空武器数据链需要充分结合航空武器数据链自身特点解决特定阵列形式和信号特性下的波达方向(Direction of Arrival,DOA)估计问题,高动态条件下对误差不敏感的稳健自适应波束形成问题以及适应于航空武器平台有限运算能力的算法简化问题。本文针对以上问题进行了深入研究,主要内容与创新点如下:1)针对非均匀线阵下非圆(Non-Circular,NC)信号DOA估计中需进行角度和非圆相位二维搜索的高复杂度问题,提出了降维NC-Capon、降维NC-MUSIC和降维NC-PM算法。所提出的降维算法利用非圆特性扩展阵列孔径,角度估计性能优于传统算法。同时算法只需进行一维搜索,具有低的复杂度,且角度估计性能非常接近于二维谱峰搜索算法。推导了非均匀线阵下非圆信号DOA估计的克拉美-罗下界(CRB),为算法理论分析提供了强有力的支撑。另外,将广义ESPRIT与非圆特性结合起来,提出了非均匀线阵下非圆信号的广义ESPRIT方法来估计DOA,推导了非均匀线阵下的基于频谱搜索NC-GESPRIT和求根NC-GESPRIT的DOA估计算法。所提出的NC-GESPRIT方法比GESPRIT方法有更好的角度估计性能,且在无需知道非圆相位信息的情况下也能很好地估计DOA。2)研究了非均匀L型阵列二维DOA估计问题。考虑到非均匀L型阵列进行二维DOA估计时需要角度二维搜索,提出了级联MUSIC、级联Capon和级联PM三种DOA估计算法。所提出级联算法采用级联的一维搜索实现二维DOA的联合估计,能避免二维搜索中高复杂度问题,计算复杂度大大降低,且性能接近于二维估计算法。同时算法能实现二维角度自动配对。3)研究了双平行线阵二维非圆DOA估计问题,推导了双平行线阵下基于NC-ESPRIT的二维DOA估计算法和基于NC-PM算法的二维DOA估计算法。二维NC-ESPRIT和二维NC-PM算法充分利用了信号的非圆特性,角度估计性能分别优于传统的ESPRIT算法和PM算法。提出了双平行阵列下一种基于NC-PARAFAC的非圆信号二维DOA估计算法,该算法利用信号的非圆特性扩展平行因子模型,使用三线性分解方法得到方向矩阵的估计值,进而得到自动配对的仰角和方位角。该算法角度估计性能好于ESPRIT算法、PM算法、PARAFAC算法。4)最后研究了基于广义特征空间的稳健自适应波束形成问题,针对航空武器数据链应用中,期望信号功率较强,并且由于航空武器平台的高动态运动特性和工程研制中的阵列误差等原因,导致存在期望信号指向误差、零陷指向误差、天线旁瓣较高等问题,将幅度加权技术引入期望信号和干扰子空间,提出了一种新型的基于特征空间的稳健LCMV自适应波束形成算法,然后将零陷加宽技术引入基于特征空间的稳健LCMV自适应波束形成算法,研究了锥化后的数据协方差矩阵特征空间分解方法,提出了具有宽零陷特性的稳健波束形成算法,且具备较好的稳健性、收敛速度和低旁瓣特性。
[Abstract]:Aviation weapon data link is used in aviation weapon target information correction, special data cooperative control chain, is the key equipment to improve aviation weapon time sensitive targets strike capability, a key to get a tactical advantage is also under conditions of informatization system. The working process of air combat weapon data link is a gradually away from one's own platform. The process of approaching enemy platform, traditional anti interference measures in the face of signal acquisition speed reaches microsecond, interference of equivalent radiated power reaches megawatt, a new generation of interference suppression platform can accurately, it will be hard to obtain satisfactory results. The anti-jamming array technology for special treatment by the received signals of the antenna array in space according to a certain arrangement, the formation of high gain main lobe in the direction of the desired signal, enhance the useful signal energy, nulling in the direction of interference, interference suppression and signal The noise for airborne weapon data link can greatly improve the anti-interference ability. Because of the particularity of the environment in aviation weapon data link needs to be fully combined with the aviation weapon data link characteristics to solve specific form of array and signal characteristics of the wave direction of the application of anti jamming Technology (Direction of Arrival, DOA array) estimation problem, simplify the problem robust adaptive beamforming under high dynamic condition of error is not sensitive to the problems and adapt to the formation of aviation weapon platform limited computing algorithm. In this paper, the thorough research in view of the above question, the main contents and innovations are as follows: 1) for non uniform linear array of non-circular (Non-Circular, NC) for angle estimation of DOA signal and the non circular phase two-dimensional search complexity was proposed for dimensionality reduction of NC-Capon, NC-MUSIC and NC-PM reduced dimensionality reduction algorithm. The proposed dimensionality reduction algorithm using non circular features Extend the array aperture, the angle estimation performance is better than the traditional algorithm. The algorithm only needs one dimension search, has low complexity, and the angle estimation performance is very close to the 2-D spectrum search algorithm was deduced. Non uniform linear array non-circular signal DOA estimation Cramer Rao bound (CRB), provides a strong the support for the theoretical analysis of algorithm. In addition, the generalized ESPRIT combined with non circular characteristics, put forward the nonhomogeneous generalized ESPRIT method of circular array signal to estimate DOA, deduced the non uniform linear array spectrum NC-GESPRIT and NC-GESPRIT search for the root DOA estimation algorithm based on NC-GESPRIT. The proposed method has the better performance of angle estimation than the GESPRIT method, and without the need to know the non-circular phase information under the condition of well estimated DOA.2) studies the problem of non-uniform L array DOA estimation. Considering the inhomogeneous type L array The column need to estimate the 2-D DOA angle two-dimensional search, proposed cascade MUSIC, cascade Capon and cascaded PM three DOA estimation algorithm. The proposed algorithm uses the cascade cascade of one-dimensional search for joint estimation of 2D DOA, can avoid the high complexity problem in two-dimensional search, the computational complexity is greatly reduced, and the performance is close to two at the same time estimation algorithm. The algorithm can achieve automatically paired two-dimensional angle.3) of two parallel linear estimation problem of two dimensional array of non circular DOA, are two parallel linear arrays based on two-dimensional DOA estimation algorithm and NC-ESPRIT estimation based on two-dimensional DOA algorithm NC-PM algorithm. The 2D NC-ESPRIT and 2D NC-PM algorithm makes full use of the non-circular property the angle estimation performance, are better than the traditional ESPRIT algorithm and PM algorithm. This paper proposes a double parallel array based on noncircular signals of two-dimensional DOA estimation algorithm for NC-PARAFAC, the Farley. Extended parallel factor model with noncircular signals, using three linear decomposition method to get the estimation of direction matrix, then get the automatic matching of elevation and azimuth angle estimation. The algorithm is better than ESPRIT algorithm, PM algorithm, PARAFAC algorithm,.4) at the end of the formation of the robust beamformer based on generalized feature space. According to the aviation weapon data link application, the desired signal power is strong, and because of the high dynamic performance and engineering development of aviation weapon platform in the array error and other reasons, leading to the existence of the desired signal pointing error, null pointing error, the antenna sidelobe problem of higher amplitude weighting technique into the desired signal and the interference subspace. We put forward a new algorithm of robust LCMV adaptive beamforming based on feature space, then the null broadening technique into robust LCMV based on feature space Adaptive beamforming algorithm is studied, and the eigenspace decomposition method of cone covariance matrix is studied. A robust beamforming algorithm with wide nulling feature is proposed, which has better robustness, convergence speed and low sidelobe characteristics.

【学位授予单位】：南京航空航天大学
【学位级别】：博士
【学位授予年份】：2016
【分类号】：E926;TN911.7

【相似文献】