Matrix inversion decryption method based on orthogonalization and least square iteration
By employing Schmitt orthogonalization and least squares iteration, the problem of incomplete filtering of redundant and noise components in traditional matrix inversion methods is solved, improving the accuracy and stability of the cryptographic matrix inverse and achieving more efficient decryption.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- BEIJING INSTITUTE OF TECHNOLOGY (ZHUHAI)
- Filing Date
- 2026-05-06
- Publication Date
- 2026-07-17
AI Technical Summary
Traditional matrix inversion methods are ineffective in handling redundant and noisy components and are sensitive to singular values in ill-conditioned matrices, resulting in insufficient decryption accuracy and stability.
By employing Schmidt orthogonalization and least squares iteration, the cryptographic matrix is transformed into a set of orthogonal unit vectors. The error is quantified by the bias matrix, and the matrix is corrected element by element by least squares iteration in both row and column dimensions, thereby improving the accuracy and stability of the inverse matrix.
It improves the accuracy and stability of solving the inverse of the cryptographic matrix, provides a faster decryption method, and reduces the impact of calculation errors and noise.
Smart Images
Figure QLYQS_1 
Figure QLYQS_2 
Figure QLYQS_3