冲击噪声环境中最小“几何功率”误差波束形成算法

Minimum "Geometric Power" Error Beamforming Amid Heavy-Tailed Impulsive Noise of Unknown Statistics

本文提出一种适用于任意未知统计特性的代数拖尾冲击噪声(包含所有对称α稳定分布噪声)环境下的波束形成算法.算法利用输出信号和参考信号之间"几何功率"误差的最小化来求解最优权向量."几何功率"误差定义成误差信号的对数矩的形式.我们采用迭代复加权最小二乘估计来求解最小"几何功率"误差波束形成权向量.与基于最小分数低阶误差波束形成算法相比,最小"几何功率"误差波束形成算法计算更为简单;不需要噪声特征指数的先验信息或估计;适用于更广的冲击噪声环境;具有更小的估计误差.计算机仿真验证了算法的有效性.

This paper proposes a new beamforming approach, against arbitrary algebraicaUy-tailed impulsive noise of otherwise unknown statistics. （This includes all symmetric alpha stable noises）. This new beamformer minimizes the “geometric power” error between the beamformer＇ s output and the reference signal. This “geometric power” error is defined in terms of the logarith- mic moment. The iteratively re-weighted least squares （IRIS） algorithm is adopted to calculate the proposed beamformer weights. Relative to costmary fractional lower order errors based beamformer, the proposed beamformer offers advantages such as： simpler computationally; needing no prior information nor estimation of the numerical value of the impulsive noise＇s effective characteristic exponent, applicable to a wider class of impulsive noises; and improving the performance in terms of lower estimation errors. Computer simulation results verify the efficacy of the proposed beamfonner.