Title :
The use of small pivot perturbation in circuit analysis
Author_Institution :
IBM Gen. Technol. Div., Hopewell Junction, NY, USA
fDate :
11/1/1991 12:00:00 AM
Abstract :
A technique that can improve the accuracy of Gaussian elimination in those cases where small pivot magnitudes are encountered during the Gaussian elimination procedure, thus avoiding an inaccurate solution and allowing the analysis to continue, is presented. This technique is known as small pivot perturbation (SPP). It is shown that SPP can be used to provide accurate solutions in those cases where Gaussian elimination without row or column pivoting has resulted in a very small or zero-valued pivot. The SPP technique has been implemented in the AS/X circuit analysis program. For the AS/X circuit analysis application, where maintenance of matrix sparsity has ruled out any pivoting during Gaussian elimination, SPP is the most practical technique yet found for dealing with small pivots
Keywords :
circuit analysis computing; matrix algebra; perturbation techniques; AS/X; Gaussian elimination; circuit analysis program; matrix sparsity; small pivot perturbation; Algorithm design and analysis; Circuit analysis; Computer errors; Design automation; Equations; Frequency domain analysis; Linear systems; Matrix decomposition; Sparse matrices; Time domain analysis;
Journal_Title :
Computer-Aided Design of Integrated Circuits and Systems, IEEE Transactions on