低先验信息依赖性的稳健波束形成技术及其应用研究

发布时间：2018-05-06 22:20

本文选题：自适应波束形成 + 稳健性　；参考：《哈尔滨工程大学》2014年博士论文

【摘要】：自适应波束形成器是阵列信号处理中实现目标检测、参数估计等的一个重要手段。为了提高阵列系统在存在条件失配和强干扰的环境中对远场目标的检测和参数估计能力,木文针对自适应波束形成器对稳健性的迫切需求,结合水声设备实际应用环境,从提高自适应波束形成系统稳健性和阵列校正技术两个方面,开展了深入的理论研究和实验分析。主要研究内容为:1.针对存在各种失配时标准Capon波束形成器性能严重下降问题,提出一种基于最小二乘估计的稳健波束形成算法,同时针对二阶锥规划计算量较大问题,提出一种计算复杂度较低的一维搜索求解方法。该算法结合了广义旁瓣对消器和稳健最小二乘估计的思想,能同时缓解导向矢量误差和快拍数不足的不利影响,通过仿真和试验验证了该算法的有效性和对人为设置参数的低敏感性。同时将该算法推广至声矢量阵,提出基于最小二乘估计的声矢量阵稳健波束形成算法,仿真分析表明该算法性能优于现存的声矢量阵稳健波束形成算法,且在抗阵元姿态误差上性能更为突出。2.针对导向矢量失配程度较大时,常见稳健自适应波束形成算法性能不足问题,提出一种基于最优导向矢量估计的稳健波束形成算法。该算法直接利用接收数据估计最优导向矢量,唯一的先验信息是期望信号可能的入射区间,该先验信息在实际应用容易获取。数值仿真和试验结果表明该算法在抗导向矢量误差和快拍数不足时的有效性,且在观测方向误差较大时,该算法的性能较常见的稳健波束形成算法有较大优势。3.针对声矢量阵阵元位置误差和姿态误差同时存在的情况,提出一种基于差分进化的声矢量阵阵型误差校正算法。以达尔文进化理论为基本思想的差分进化算法,在求解高维非线性优化问题时有独特的优势。新算法以加权子空间拟合的目标函数为代价函数,引入差分进化思想求解。该算法只需待估参数误差的上下界这一先验信息,即可实现对阵元位置误差和姿态误差的联合校正。数值仿真分析验证了该算法的有效性。4.基于相干信号的处理方法是实现宽带信号参数估计重要方法,具有技术复杂度低、能实现对相干信号处理等优点。但将其应用于宽带自适应波束形成时,宽带聚焦误差和期望信号预估方位的偏差都会导致后续自适应波束形成算法的约束条件出现失配。针对上述问题,提出一种适用于声矢量阵的高稳健性宽带聚焦波束形成方法。该算法先为声矢量阵设计了一种聚焦误差较低的宽带聚焦方法,并在后续自适应波束形成中采用稳健的处理思想。仿真分析表明,该算法不仅具有计算复杂度低、束宽恒定的优势,且在抗观测方向误差上表现出较高稳健性。
[Abstract]:Adaptive beamformer is an important method for target detection and parameter estimation in array signal processing. In order to improve the ability of far field target detection and parameter estimation in the presence of conditional mismatch and strong interference, the paper aims at the urgent need for robustness of adaptive beamformer and combines the practical application environment of underwater acoustic equipment. In order to improve the robustness of adaptive beamforming system and array correction technology, the theoretical research and experimental analysis are carried out. The main research contents are: 1: 1. A robust beamforming algorithm based on least square estimation is proposed to solve the problem of serious deterioration of performance of standard Capon beamformer with various mismatch times. A one-dimensional search method with low computational complexity is proposed. The algorithm combines the idea of generalized sidelobe canceller and robust least square estimation, which can alleviate the adverse effects of the error of guide vector and the insufficient number of beats simultaneously. The effectiveness of the algorithm and its low sensitivity to artificial setting parameters are verified by simulation and experiments. At the same time, the algorithm is extended to acoustic vector array, and a robust beamforming algorithm based on least square estimation is proposed. The simulation results show that the proposed algorithm is superior to the existing robust beamforming algorithm. And the performance of anti-array element attitude error is more prominent. 2. A robust beamforming algorithm based on optimal guidance vector estimation is proposed to solve the problem of poor performance of conventional robust adaptive beamforming algorithms when the mismatch of steering vector is large. The algorithm directly uses the received data to estimate the optimal guidance vector, and the only prior information is the possible incidence interval of the desired signal, which is easy to obtain in practical applications. The numerical simulation and experimental results show that the algorithm is effective when the steering vector error and the number of beats are insufficient, and the performance of the algorithm is better than that of the conventional robust beamforming algorithm when the error of observation direction is large. Aiming at the simultaneous errors of position and attitude of acoustic vector array, a differential evolution based correction algorithm for acoustic vector array error is proposed. The differential evolutionary algorithm based on Darwin's evolution theory has a unique advantage in solving high dimensional nonlinear optimization problems. The new algorithm takes the objective function of weighted subspace fitting as the cost function and introduces the idea of differential evolution to solve the problem. This algorithm only needs the priori information of upper and lower bounds of parameter error to be estimated to realize the joint correction of position error and attitude error. The effectiveness of the algorithm is verified by numerical simulation. 4. 4. The processing method based on coherent signal is an important method for parameter estimation of wideband signal, which has the advantages of low technical complexity and the ability to process coherent signal. However, when applied to wideband adaptive beamforming, the deviation of wideband focusing error and expected signal estimation azimuth will lead to the mismatch of the constraint conditions of the subsequent adaptive beamforming algorithm. To solve the above problems, a high robust wideband focusing beamforming method for acoustic vector arrays is proposed. Firstly, a wideband focusing method with low focusing error is designed for acoustic vector array, and the robust processing idea is adopted in subsequent adaptive beamforming. Simulation results show that the proposed algorithm not only has the advantages of low computational complexity and constant beam width, but also has high robustness in resisting observation direction errors.
【学位授予单位】：哈尔滨工程大学
【学位级别】：博士
【学位授予年份】：2014
【分类号】：TN911.7

【相似文献】