小波多分辨率分析改进互连结构电容提取

Capacitance Extraction of Interconnects by the Wavelet Multiresolution Decomposition

本文将点匹配矢量法矩阵视作类似离散图像信号的一组数字信号矢量，进行小波多分辨率分解对该矩阵进行拟对角化压缩，克服了以往小波基直接展开的矩量法在求解二维与三维电磁问题时的困难．文章例子的计算结果说明了这种方法在将矩量法得到的满矩阵变换为稀疏矩阵时具有较高的效率．电容参数提取的效率也由于矩阵的稀疏化得到了提高．文中的数据值结果与文献及ＦＡＳＴＣＡＰ软件的结果吻合．

Instead of the direct expansion of the charge distribution by the orthogonal wavelet basis,the large full matrix gained from MoM (Method of Moments) discretization of the integral equations is taken as discrete digital signals and sparsified by two dimensional multiresolution representation.Thus,the difficulty of using wavelets on the two and three dimensional structures is overcome.Computational results are given for several typical interconnects and the data obtained show that the proposed method substantially sparsifies the matrix equation.The capacitance values calculated by the matrix equations with high sparsity agree well with the results of the previous papers and that of capacitance extractor FASTCAP.