首页> 外国专利> The eigenvalue decomposition and The Singular Value decomposition of matrices using Jacobi rotation

The eigenvalue decomposition and The Singular Value decomposition of matrices using Jacobi rotation

机译:利用Jacobi旋转的矩阵特征值分解和奇异值分解

摘要

Techniques for parsing using the Jacobi rotation matrices. Multiple iterations of Jacobi rotation are conducted on a First array of Complex values of Jacobi matrices with multiple rotation of Complex values to Zero elements that are outside of the diagonal in the Matrix.For each iteration, one can form a subarray based on the first Matrix and can be Decomposed to obtain eigenvectors for the subarray, and can form a rotation matrix with the eigenvectors of Jacobi and can be used to update the first Matrix.A Second Matrix Complex values, which contains vectors orthogonal is derived based on the Jacobi rotation matrices. For The decomposition of the own value can derive a Third Matrix of values based on the rotation of Jacobi matrices.For The Singular Value decomposition of matrix, a fourth and a Left singular Vectors and Singular Value Matrix can be derived based on the Jacobi rotation matrices.
机译:使用Jacobi旋转矩阵进行解析的技术。在Jacobi矩阵的第一个复数值数组上进行Jacobi旋转的多次迭代,将Matrix的复数值多次旋转到零元素(位于矩阵对角线之外),每次迭代都可以基于第一个Matrix形成子数组可以分解为子数组的特征向量,并可以与Jacobi的特征向量形成旋转矩阵,并可以用于更新第一个矩阵。第二个Matrix Complex值包含基于Jacobi旋转矩阵的正交向量。对于自身值的分解,可以基于Jacobi矩阵的旋转得出值的第三矩阵;对于矩阵的奇异值的分解,可以基于Jacobi旋转矩阵得出第四和左奇异矢量和奇异值矩阵。

著录项

获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号