Method for correcting reciprocity error in two port network measurements
US6647357B1 · kind B1 · utility
Assignee
Inventor
Key dates
| Filing date | Feb 7, 2000 |
| Grant date | Nov 11, 2003 |
| Priority date | — |
| Expiry date | Feb 7, 2020 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG01R27/28
- WIPO fieldMeasurement
- WIPO sectorInstruments
Abstract
A method for correcting reciprocity error in two port network measurements, in which an iterative algorithm is used to obtain an optimal symmetric matrix approximation to a measured 2×2 reciprocal network. The algorithm smooths measured data to obtain a low noise floor on measurements related to mode conversion in cables and connecting hardware. A geometric interpretation is used to approximate a measured data vector with the closest vector which satisfies the reciprocity constraint. An initial point in the constraint surface (reciprocity manifold) is located, and another point which is a better approximation of the measured data vector which satisfies the reciprocity constraint is generated. By repeating this process, the closest point in the manifold is quickly located. This closest point defines a reciprocal matrix which can be used to estimate the actual device under test (DUT) parameters.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.