利用导向向量旋转和联合迭代优化的自适应波束形成算法研究  

Adaptive beamforming algorithm based on direction vector rotation and joint iterative optimization

作　　者：谢建平[1] 王瑞[2] 何熊熊[2] 李胜[2] 

机构地区：[1]丽水学院工程与设计学院,丽水323000 [2]浙江工业大学信息工程学院,杭州310023

出　　处：《声学学报》2016年第3期343-352,共10页Acta Acustica

基　　金：国家科技支撑计划课题(2013BAF07B03);浙江省自然科学基金(LY13F010009)资助

摘　　要：针对大型阵列中自适应波束形成技术的实时性和鲁棒性问题,基于最小方差无失真响应(Minimum Variance Distortionless Response,MVDR)波束形成的信号模型框架,提出一种通过对导向矢量进行处理以降低干扰的自适应波束形成算法——稳健联合迭代优化-导向自适应(Robust Joint Iterative Optimization-Direction Adaptive,RJIO-DA)算法。在联合迭代优化的基础上,将降维变换矩阵的每一个列向量看作独立的方向向量,引导子空间内每一个维度上的权值迭代,同时旋转导向向量,减小了由于导向误差的不确定性而导致的性能下降。仿真实验结果表明,与现有的降维算法相比,RJIO-DA算法计算复杂度低、收敛率高、鲁棒性好,可在期望方向上稳健地聚集波束,更好地形成干扰方向的自适应零陷。A robust reduced-rank algorithm named RJIO-DA （Robust Joint Iterative Optimization-Direction Adaptive） is proposed for adaptive beamforming in large array scenarios. Based on a MVDR （Minimum Variance Distortionless Response） framework, the proposed algorithm jointly updates a transforming matrix and a reduced-rank filter. In order to reduce the complexity and improve the performance, each column of the transforming matrix is regarded as a direction vector on each dimension of the subspace and is updated individually. In addition, the limited uncertainties caused by the direction error can be overcome with the proposed algorithm. The simulation results show that RJIO-DA algorithm has lower complexity and faster convergence compared with the conventional reduced-rank algorithms.

关 键 词：自适应波束形成 导向误差 降维算法 变换矩阵 导向矢量 子空间 干扰方向 算法设计 迭代优化 最小方差 

分 类 号：TN911.7[电子电信—通信与信息系统]