Characterization of Receiver Non-Linearity in Vector Network Analyzer Measurements
Abstract
Accurate scattering (S)-parameter measurements using vector network analyzers (VNAs) are essential in RF and microwave applications. While calibration procedures correct for most systematic errors and aim to ensure traceability, non-linearity in VNA receivers might not be the most prominent but remains a significant and complex source of uncertainty to investigate. In this work we examine the VNAs linearity without the need of known or calibrated artifacts. To achieve this target, a linear artifact (e.g. an attenuator) is measured at different source powers and the receiver characteristics of the a- and b-receivers are determined based on a model using these measurements.
Full text
Characterization of Receiver Non-Linearity in Vector Network Analyzer Measurements Mohanad Dawood Al-Dabbagh, David Ulm, and Thomas Kleine-Ostmann Physikalisch-Technische Bundesanstalt (PTB), Department 2.2 - High Frequency and Electromagnetic Fields, Bundesallee 100, 38116 Braunschweig Commission A Accurate scattering (S)-parameter measurements using vector network analyzers (VNAs) are essential in RF and microwave applications. While calibration procedures correct for most systematic errors and aim to ensure traceability, non-linearity in VNA receivers might not be the most prominent but remains a significant and complex source of uncertainty to investigate. Non-linearity often arises in receiver components such as amplifiers and mixers, where it cannot be adequately corrected using standard calibration models. As a result, its impact is typically quantified as an additional uncertainty in the corrected S-parameters. It contributes to the extended uncertainty error terms of VNAs, alongside noise, drift, cable movements and connector repeatability. Existing methods for evaluating non-linearity, such as precision artifact-based techniques, are limited to specialized metrology laboratories and are typically applicable only within restricted frequency ranges. In this work we examine the VNAs linearity without the need of known or calibrated artifacts. To achieve this target, a linear artifact (e.g. an attenuator) is measured at different source powers and the receiver characteristics of the aand b-receivers are determined based on a model using these measurements. This approach is validated in the low frequency range of the original frequency range of the VNA. For this purpose, the receiver characteristics are also determined using a highly linear thermal power sensor and both measurements are compared with each other. The source power was tuned using an external attenuation once, and by performing a power sweep. The measurement investigation is then extended to the mm-wave range. Here, we applied multiple attenuation levels to both single-port and two-port configurations to observe powerdependent deviations in the measured wave quantities of the receiver characteristics using reflection and transmission measurements and compared with each other. Finally, we conducted over-the-air (OTA) measurements incorporating a quasi-optical measurement testbed and with the use of polarizers at the collimated beam region to investigate angular and polarization dependencies of receiver non-linearity in spatial test setups. The results of this investigation will highlight the influence of receiver non-linearity on traceable VNA measurements and demonstrate the need for improved uncertainty models that incorporate non-linear receiver effects, particularly in wideband and high-power applications. The paper was presented during the Kleinheubacher Tagung 2025 held on 23 to 25 September 2025 in Miltenberg, Germany. For more details see https://www.kh2025.de/