An Implementation of Matrix Eigenvalue Decomposition with Improved Jacobi Algorithm
Wei Mei, Ming Jin, Shuai Liu, Qiao Xiao Lin, Qian Wei Qiang
Abstract
Wei Mei, Ming Jin, Shuai Liu, Qiao Xiao Lin, Qian Wei Qiang
Abstract
Eigenvalue decomposition for real symmetric matrix is significant in mathematics and engineering. In engineering implementation, most of implementation for eigenvalue decomposition based on hardware prefers to choose Jacobi algorithm because of its inherent parallelism. But the calculated eigenvalue and its corresponding eigenvector from traditional Jacobi algorithm are unordered arrangement. To solve this problem, an improved Jacobi is proposed in this paper, which can get eigenvalue and eigenvector in descending order.
OpenAlex reports 5 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Eigenvalue decomposition for real symmetric matrix is significant in mathematics and engineering. In engineering implementation, most of implementation for eigenvalue decomposition based on hardware prefers to choose Jacobi algorithm because of its inherent parallelism. But the calculated eigenvalue and its corresponding eigenvector from traditional Jacobi algorithm are unordered arrangement. To solve this problem, an improved Jacobi is proposed in this paper, which can get eigenvalue and eigenvector in descending order.
Key concepts: Jacobi eigenvalue algorithm, Eigenvalues and eigenvectors, Divide-and-conquer eigenvalue algorithm, Eigendecomposition of a matrix, Jacobi method, Inverse iteration, Algorithm, Decomposition