Covariance-Based Vector-Network-Analyzer Uncertainty Analysis for Time- and Frequency-Domain Measurements

2009 74th ARFTG Microwave Measurement Conference

2009

Self Cite

Abstract-We propose a new description of connector repeatability errors for coaxial one-port devices. Our approach is based on a stochastic model constructed as a lumped-element equivalent circuit with randomly varying frequency-independent parameters. We represent statistical properties of these parameters with a covariance matrix which is estimated from a small number of repeated measurements (typically 16) of a one-port device under test. We illustrate our approach by modeling connector repeatability errors for 1.85 mm coaxial offset shorts. These results show that our model is capable of reproducing the complicated frequencydependent behavior of connector repeatability errors for coaxial one-port devices with typically only two or three random parameters.

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Stochastic modeling of coaxial-connector repeatability errors

Lewandowski

2009 74th ARFTG Microwave Measurement Conference

2009

Self Cite

“…The NIST Microwave Uncertainty Framework utilizes parallel sensitivity and Monte-Carlo analyses, and enables us to capture and propagate the significant Sparameter measurement uncertainties and statistical correlations between them [11]. By identifying and modeling the physical error mechanisms in the calibration standards, we can determine the statistical correlations between both the scattering parameters at a single frequency and uncertainties at different frequencies.…”

Section: Introductionmentioning

confidence: 99%

Developing models for Type-N coaxial VNA calibration kits within the NIST Microwave Uncertainty Framework

Jargon

2016 87th ARFTG Microwave Measurement Conference (ARFTG)

Hale

2016

-We developed models for Type-N coaxial vector network analyzer (VNA) calibration kits within the NIST Microwave Uncertainty Framework. First, we created physical models of commercially-available standards that support multiline thru-reflect-line (TRL) and open-short-load-thru (OSLT) calibrations, and included error mechanisms in each of the standards' constituent parameters that were utilized in the NIST Microwave Uncertainty Framework to propagate uncertainties. Next, we created a measurement-based model of a commercial electronic calibration unit (ECU) by characterizing the scattering parameters of its internal states with a multiline TRL calibration. Finally, we calibrated a network analyzer using the three calibration methods, and compared measurements, including uncertainties, made on a number of verification devices. We show that the three calibrations agreed to within their respective uncertainties.

“…The NIST Microwave Uncertainty Framework utilizes parallel sensitivity and Monte-Carlo analyses, and enables us to capture and propagate the significant S-parameter measurement uncertainties and statistical correlations between them [7]. By identifying and modeling the physical error mechanisms in the calibration standards, we can determine the statistical correlations between both the scattering parameters at a single frequency and uncertainties at different frequencies.…”

Section: Introductionmentioning

confidence: 99%

Physical models for 2.4 mm and 3.5 mm coaxial VNA calibration kits developed within the NIST microwave uncertainty framework

Jargon

Cho

2015 85th Microwave Measurement Conference (ARFTG)

et al. 2015

-We developed physical models of commerciallyavailable 2.4 mm and 3.5 mm coaxial calibration kits for vector network analyzers. These models support multiline thru-reflectline (TRL) and open-short-load-thru (OSLT) calibrations, and include error mechanisms in each of the standards' constituent parameters that can be utilized in the NIST Microwave Uncertainty Framework to propagate uncertainties. For both connector sizes, we calibrated a network analyzer using the two calibration methods, and compared measurements and uncertainties made on a number of verification devices. In both cases, we showed that the two calibrations agree to within their respective uncertainties.