Using the Variable-Length Arithmetic for an Accurate Poles-Zeros Analysis
In the paper, a reduction algorithm for transforming the generaleigenvalue problem to the standard one is presented for both classicalfull-matrix methods and a sparse-matrix technique appropriate forlarge-scale circuits. An optimal pivoting strategy for the two methodsis proposed to increase the pre...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Spolecnost pro radioelektronicke inzenyrstvi
2003-09-01
|
| Series: | Radioengineering |
| Online Access: | http://www.radioeng.cz/fulltexts/2003/03_03_01_05.pdf |
| Summary: | In the paper, a reduction algorithm for transforming the generaleigenvalue problem to the standard one is presented for both classicalfull-matrix methods and a sparse-matrix technique appropriate forlarge-scale circuits. An optimal pivoting strategy for the two methodsis proposed to increase the precision of the computations. The accuracyof the algorithms is furthermore increased using longer numerical data.First, a ORQJ.GRXEOH precision sparse algorithm is compared with theGRXEOH precision sparse and full-matrix ones. Finally, the applicationof a suitable multiple-precision arithmetic library is evaluated. |
|---|---|
| ISSN: | 1210-2512 |