Recent Study Reveals Dramatic Differences in Chipset and Smartphone Location Performance

Signals Research Group (SRG) has collaborated with Spirent several times in the past on chipset and device performance benchmarking and the results are always eagerly anticipated by my subscribers – a cross-section of leaders in the wireless industry. This time, we benchmarked the performance of various A-GNSS chipset platforms and their implementation in a number of the most popular smartphones on the market today. Spirent provided its 8100 Location Technology platform and SRG was responsible for analyzing the data and publishing the results of the study. 

Going into the study, I expected that there would be some measurable performance differences, but I had no idea that the differences would be so large and the implications so important. Although the performance of the underlying chipset platforms was largely comparable, we observed dramatic differences in the over-the-air (OTA) testing of the smartphones in an anechoic chamber.

The most startling finding was that the introduction of a protective case can have a material impact on the receive sensitivity of the handset + A-GNSS chipset configuration. One protective case, in particular, lowered the receive sensitivity by 8 dB, and when combined with the poor performance of the smartphone it made the A-GNSS solution essentially unusable in many real-world situations. Since I happen to be the unfortunate owner of this smartphone, I can personally attest that this poor performance can be observed outside of the lab environment. Another protective case that we tested had virtually no impact on the receive sensitivity but it did alter the radiated antenna pattern that we observed.

Nonetheless, operators may need to reconsider how they perform handset acceptance testing. It is doubtful that operators include protective cases when they conduct their acceptance testing and they may not have even considered how a protective case could influence the performance of the handset. In many cases – no pun intended – the protective cases are sold in the aftermarket or they are supplied by third parties that have no affiliation with the handset manufacturers. Still, the onus is on the operator, not to mention the handset manufacturer, to ensure that any accessory that is sold with or for a particular smartphone model does not negatively impact its performance. This philosophy is especially important given the multi-billion dollar opportunity for location based services and the stringent E911 requirements that exist in the United States, with similar, albeit less stringent, requirements in other regions of the world.

Even without the protective case there are meaningful performance differences in the commercially implemented A-GNSS solutions that we tested – differences that could only be observed through careful testing in a controlled environment. Of the ten smartphones that we tested we found an 11.8 dB difference in the peak receive sensitivity between the best performing smartphone and the worst performing smartphone. Since we only sampled one device per model, I can’t necessarily extend the results to all smartphones within a particular model, but given all of the retesting that was done, the results are definitely accurate for the specific devices that we tested. It also appears that the performance differences have more to do with the particular design of the smartphone, the selection of the A-GNSS antenna and its location within the smartphone than capabilities of the underlying chipset. 

The good news, if you will, is that the soon-to-be introduced integration of A-GPS and A-GLONASS should result in an overall improvement in the A-GNSS performance with higher yields, a faster time to first fix, and higher accuracy in the geo location fix. We tested two vendor supplied A-GNSS platforms that support both A-GPS and A-GLONASS capabilities and found that in more challenging conditions, such as in urban canyons where the number of observable GPS satellites could be limited, that the added benefit of the GLONASS satellites can increase the yield by up to a factor of 4x. Obviously, in some cases, such as when there are already numerous observable GPS satellites, the incremental benefits of GLONASS are negligible. 

If you are interested in the published report, please feel free to visit our website at www.signalsresearch.com where you can download a report preview.

comments powered by Disqus
× Spirent.com uses cookies to enhance and streamline your experience. By continuing to browse our site, you are agreeing to the use of cookies.