Publications Confirm: DASY6 is the Most Accurate SAR System

A recent publication implied that probe-scanning SAR measurement systems have large measurement uncertainties (up to 40%). This does not apply for the DASY systems as demonstrated by the IT’IS Foundation who found that the maximum DASY error is less than 3%. The results were published and confirmed by the authors of the original study.

The article, “Discrepancies of Measured SAR between Traditional and Fast Measuring Systems”, by Zicheng Liu, Djamel Allal, Maurice Cox, and Joe Wiart was published in March 2020 in International Journal of Environmental Research and Public Health. The work is part of an EU-funded project initiated by ART-Fi SAS. The objective of the paper was to analyze the sources of post-processing error of two types of specific absorption rate (SAR) measurement systems: time-domain array systems and probe-scanning systems. The post-processing errors were analyzed using eleven source antennas that are described in IEC 62209-3 Annex H.

For the time-domain array systems, the authors analyzed the post-processors errors applying ideal measurement systems without measurement errors. The results showed that the problem of reconstruction is ill-conditioned, especially if the sensors are in larger distance to the phantom surface (>10 mm). This confirms an early analysis that was performed by our research partner, the IT’IS Foundation, that led to SPEAG's decision to have minimal sensor-to-surface distance in cSAR3D of 3 mm.

The article also analyzed the post-processing errors of probe-scanning systems. Unfortunately, the authors used scanning and evaluation parameters that are not compliant with the requirements of the published standards. This led to incorrect results and the misleading conclusion that these errors can be as much as 40%.

We mandated IT’IS with repeating the analysis applied in the paper with state-of-the art scanning and evaluation parameters of the DASY technology. The analysis was performed in collaboration with two authors of the original study using the same raw data. The results demonstrate that the maximal post-processing error is less than 3% when applying the recommended parameters in DASY6 using the EX3DVx probes. The findings were published as a Comment in the same journal. The authors of the original study confirmed these findings in their Reply paper.

Since SPEAG substantially invests in research of the electromagnetic near fields, it can react and adopt the latest findings earlier than anybody else. For example, when the research showed that extra-polation to the surface is the biggest error source for very localized exposures, SPEAG developed and released the EX3DVx probe technology in 2003 that enables measurements as close as 1.4 mm. Similarly, SPEAG developed the technology allowing the placement of the sensor array in cSAR3D very close to the surface.