阵列信号处理中稳健自适应波束形成算法研究

发布时间：2018-08-31 10:57

【摘要】：阵列信号处理具有波束控制灵活、信号增益高、干扰抑制能力强等优点,自适应波束形成是阵列信号处理中的重要研究方向,通过自适应的调整权值使阵列方向图主瓣指向期望信号方向,零点对准干扰方向,提高输出信噪比。在实际应用中,由于存在离散扫描间隔和阵元误差失配造成的导向向量误差以及有限采样快拍数造成数据协方差矩阵估计误差,因此,研究具有误差鲁棒性的稳健自适应波束形成算法具有重要研究意义。本文首先建立阵列信号处理数学模型,介绍自适应波束形成算法基本思想。然后引入三种最优波束形成准则(MMSE、MSINR、LCMV)和几种经典的自适应算法(LMS、RLS、GSC)。为了克服经典自适应算法对模型误差敏感的缺点,论文引入了在模型失配下仍能保证良好输出性能的稳健自适应算法,介绍了LSMI、ESB、RCB三种经典的稳健自适应算法,并分析其性能优缺点。论文针对当采样快拍数据含有期望信号分量时,现有一些算法性能衰落的缺点,提出了两种基于协方差矩阵重构的稳健自适应算法。第一种算法采用联合算法的思想,先利用Music空间谱方法重构干扰噪声协方差矩阵,去除采样矩阵中期望信号分量,再通过求解优化问题修正期望信号导向向量。仿真实验结果表明,该算法对低快拍误差具有稳健性,且增强了干扰抑制能力。另一种算法是对正交投影算法(OP)的改进,将利用Music空间谱估计方法重构干扰噪声协方差矩阵应用到正交投影算法中。仿真实验结果表明,改进OP算法解决了原OP算法在样本数据含有期望信号分量时的信号相消问题,且减小了噪声扰动对算法性能的影响,增强了干扰抑制能力。
[Abstract]:Array signal processing has the advantages of flexible beam control, high signal gain and strong interference suppression ability. Adaptive beamforming is an important research direction in array signal processing. The main lobe of the array pattern is directed towards the desired signal direction and the zero point is aligned to the interference direction by adjusting the weights adaptively to improve the output signal-to-noise ratio (SNR). In practical application, the error of covariance matrix estimation is caused by the mismatch between discrete scan interval and element error, and because of the error of guide vector caused by the mismatch of discrete scanning interval and error of array element, and the estimation error of data covariance matrix caused by the limited sampling beat number. It is important to study robust adaptive beamforming algorithm with error robustness. In this paper, the mathematical model of array signal processing is established, and the basic idea of adaptive beamforming algorithm is introduced. Then three kinds of optimal beamforming criteria (MMSE,MSINR,LCMV) and several classical adaptive algorithms (LMS,RLS,GSC) are introduced. In order to overcome the shortcoming that classical adaptive algorithm is sensitive to model error, this paper introduces robust adaptive algorithm which can guarantee good output performance under model mismatch, and introduces three classical robust adaptive algorithms of LSMI,ESB,RCB. The advantages and disadvantages of the performance are analyzed. In this paper, two robust adaptive algorithms based on covariance matrix reconstruction are proposed to overcome the shortcomings of the performance fading of some existing algorithms when the sampled rapid-beat data contains the desired signal component. The first algorithm adopts the idea of joint algorithm. Firstly, the interference noise covariance matrix is reconstructed by Music spatial spectrum method, and the desired signal component is removed from the sampling matrix, and then the desired signal guidance vector is corrected by solving the optimization problem. The simulation results show that the algorithm is robust to low beat error and enhances the ability of interference suppression. The other algorithm is to improve the orthogonal projection algorithm (OP), which uses the Music space spectrum estimation method to reconstruct the interference noise covariance matrix in the orthogonal projection algorithm. The simulation results show that the improved OP algorithm solves the signal cancellation problem of the original OP algorithm when the sample data contains the desired signal component, reduces the influence of noise disturbance on the performance of the algorithm, and enhances the ability of interference suppression.
【学位授予单位】：云南大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TN911.7
，

本文编号：2214773