Limitations of using L5 on phones

[Sean Barbeau]

Hi all,

This session in ION this week is a good explanation of why L5 hasn't performed as well as initially expected on phones. I don't know if there will be a public version of the presentation or paper.

Nav Data bits for Increasing Sensitivity and Robustness in Mass Market Receivers

Frank van Diggelen, Google; Tim Murphy, Boeing; and John Betz, The MITRE Corporation
Date/Time: Wednesday, Sep. 18, 9:20 a.m.

Modern GNSS signals (like L5) provide higher power than legacy signals, but that power is divided into pilot and data components. Tracking only the pilot components forgoes part of this enhanced signal power. Mass market receivers (e.g., phones, watches) are disadvantaged by small, inefficient receive antennas, so would greatly benefit from coherent tracking of pilot and data components, and this requires removing (“wiping”) the data. But the small, inefficient receive antennas produce signal levels too low to reliably estimate and wipe the data modulation. So, we have a conundrum: because of the weak signals, the receivers cannot observe the data symbols needed to boost the weak signals. Higher end receivers that are operating in challenging environments such as in the presence of RF interference, or with signal blockage (trees, etc.), would also benefit from the enhanced signal power and larger processing gain enabled by wiping the data modulation. This paper demonstrates the benefits of coherently combining measurements on pilot and data components, and describes a proposed method to provide the needed information over the internet.

https://www.ion.org/gnss/abstracts.cfm?paperID=13664

Sean

[Kimo Crossman]

Wow that looks interesting!

--
You received this message because you are subscribed to the Google Groups "GPSTest" group.
To unsubscribe from this group and stop receiving emails from it, send an email to gpstest_andro...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/gpstest_android/4d3a9643-51ce-4e65-bc23-0f8864273d20n%40googlegroups.com.

[Baard covington]

Can someone translate that into laymans terms?
In other words L5 needs a suitable antenna. But what sort of antenna and what is on the market today that would enable an L5 receiver to work properly? Are there any examples?

To view this discussion on the web visit https://groups.google.com/d/msgid/gpstest_android/CALV1V4_XrX6ENKV6yUp5NSw7q%2Bs5redPugTDWXMa2hsO6kqHsg%40mail.gmail.com.

[Sean Barbeau (xWF)]

tl;dr is that an internet-service to provide data symbols is needed so the device doesn't have to observe these data symbols itself (kind of another version of assisted GPS). This will allow devices with current L5 antennas to perform much better.

Sean

[V. Kelly Bellis]

Sean - Are you in Baltimore this week? Will you be there for Frank's 20-minute talk ? In any case, let's hope there's more from Frank van Diggelen.

[Sean Barbeau (xWF)]

No, unfortunately I won't be at ION this year.

Sean

--
You received this message because you are subscribed to a topic in the Google Groups "GPSTest" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/gpstest_android/Q6Zwa_mEeNY/unsubscribe.
To unsubscribe from this group and all its topics, send an email to gpstest_andro...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/gpstest_android/d29d3b85-5420-4a2c-9939-5f04015f2047n%40googlegroups.com.

[Invalid Data]

I'm trying to list smartphone's limitation.

Hardware: Small and low gain antenna, limited by size and geometry.

Integrated circuit design, which perform relatively poorer than dedicated circuit of high accuracy device.(https://satellite-navigation.springeropen.com/articles/10.1186/s43020-024-00146-5/figures/10)

Scarce radio frequency design. L2, B3I, E6, L6 is uncommon that only 2 phone support L2 to this day (while Qualcomm chips do have L2 ability).

L2 could improve robustness, L6 and E6 potentially are the augmentation signal to achieve decimeter accuracy.

Software: No augmentation or correction system built-in. Clock, orbit and ionosphere information help accuracy even if antenna is poor.

Battery saving design in favor of low update rate, less signal components tracked, less processing load.

Multipath effect is still challenging, there are research on improving multipath effect but in city valley and forest it's still a major source of inaccuracy.

Relying on INS system could decrease the global accuracy. If user is on a unpredictable heading and speed. Imagine the GNSS accuracy is 10 meters, if you Walk for 1-3 meter, you phone have to guess the next position.

[Invalid Data]

My observation of smart phone industry is, hardware design changes are not favorite of manufacturer. Changes only happen when necessary.

So the software and chip ability, which don't cost extra, is what we could expecting for. 

Scenario 1: Chip maker had design a SoC to receive augmentation signal.

And phone adopted such chip. This is what happened with Mediatek 9200/9300 SoC. Unfortunately Mediatek only developed china's BSD augmentation signal. It's yet to determine that when and wether this company develop other augmentation ability (Galileo and QZSS).

Scenario 2: A phone manufacturer integrated a positioning software and web service for PPP/PPP-RTK/RTK. And it's free for users. As I knew, Xiaomi once did that in their flagship lineup (Xiaomi 12? https://www.gpsworld.com/qualcomm-and-xiaomi-demonstrate-mobile-meter-level-positioning-capabilities/). But they seem to have discontinued such level of service in newer phone. 

Xiaomi did that again for BDS PPP web service with K60 Ultra in china market. It's a alternative to Mediatek BDS PPP augmentation. They're different approach but achieve same goal , Mediatek is generic ability to receive satellite augmentation signal. Xiaomi did that by receive information from web. 

Again for various satellite system, it's not clear when we get to see similar web service in smartphone. I think time will tell.