【摘要】 提出了一种针对有噪声频率响应的基追踪去噪的高效向量匹配(BPDN-VF)算法,该算法在向量匹配(VF)算法的基础上,利用压缩感知领域的基追踪去噪(Basis Pursuit Denoising,BPDN)方法,可以有效地从有噪声的频率响应中精确恢复传递函数.与基于方差加权的向量匹配算法相比,提出的方法不仅可以减小至少87.5%的采样次数,还可以将从有噪声的频率响应中得到的拟合结果的误差平均减少20.5%.更多还原

【Abstract】 In this paper,we propose an efficient Basis-Pursuit-Denoising-based Vector Fitting method for noisy frequency responses.On the basis of VF method,we employ the well-known Basis Pursuit Denoising(BPDN)method from the community of compressive sensing to accurately recover the transfer functions from the noisy frequency responses.Compared with the Variance Weighted Relaxed Vector Fitting method,the proposed method requires at least 87.5%fewer measurements and can achieve results with 20.5% error reduction at average from the noisy frequency responses.更多还原