L5 Data cuts out - Pixel 4

501 views
Skip to first unread message

and...@navmatic.io

unread,
Dec 1, 2020, 12:54:50 PM12/1/20
to GPSTest
Hello!
I was collecting some static data with a Pixel 4 device and noticed that the L5 data is intermittent. I was wondering if this is a common issue and if there are ways to collect continuous L5 data.

Thanks!
Andrew Wu  


Graphs from the GNSS Analysis tool

All Signals SNR
all_signals_snr.png

L1 SNR
L1_snr.png
L5 SNR
L5_snr.png
I notice the clock continuity seems to be correlated with L5 signal availability. Maybe that's a clue?

andras.l...@gmail.com

unread,
Dec 1, 2020, 1:53:04 PM12/1/20
to GPSTest
Hi Andrew,

I suppose something is wrong in your Pixel4, maybe it needs a restart or it is a ROM bug (then You cant do nothing)

Here it is my L5 SNR sheet from a short tracklog with my Xiaomi MI8 :



BR,
Andras

andras.l...@gmail.com

unread,
Dec 1, 2020, 1:55:03 PM12/1/20
to GPSTest
i can't attach pic somewhy...

here is a weblink: https://ibb.co/W0CBp4L

BR,
Andras

Sean Barbeau

unread,
Dec 1, 2020, 2:17:47 PM12/1/20
to andras.l...@gmail.com, GPSTest
Andrew,
Do you have the Android system Developer Option for "Force full GNSS measurements" enabled?

Sean


--
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/48205274-32df-4c26-a3a9-1f155007ad0an%40googlegroups.com.

and...@navmatic.io

unread,
Dec 1, 2020, 7:16:12 PM12/1/20
to GPSTest
Aha, I don't!

I'll run another test with that setting on and report back. Thanks

Sean Barbeau

unread,
Dec 17, 2020, 2:42:00 PM12/17/20
to GPSTest
Andrew,
Did you ever test this again?

I've actually been meaning to do a short blog post about the "Force full GNSS measurements" feature as it's not well documented, and having a before and after image like the one you shared above (assuming this is the problem) would be awesome :).

Thanks,
Sean

and...@navmatic.io

unread,
Dec 18, 2020, 9:12:39 PM12/18/20
to GPSTest
I haven't yet, but I'll try to get a clean log this soon.

and...@navmatic.io

unread,
Dec 24, 2020, 5:31:12 PM12/24/20
to GPSTest
I wasn't able to run this through the analysis before the holiday break, but I captured this log: https://drive.google.com/file/d/11_92Fw5ItLSxYIraVZY0GinuwhAW1qpH/view?usp=sharing

I'll run it through (or cap a new one if it's not enough) in the new year

andras.l...@gmail.com

unread,
Dec 24, 2020, 5:41:03 PM12/24/20
to GPSTest
Hi Andrew,

Something is not with this log, i got the following error message during the loading:
Analyzing log file ...
Number of satellites with Bad Measurements = 29 : 21 with FullBiasNanos not available
No epoch with valid pseudorange in ProcessGnssMeas
No valid measurements to process or plot


BR,
Andras 

and...@navmatic.io

unread,
Jan 11, 2021, 9:42:02 PM1/11/21
to GPSTest
Ok, got a chance to collect a cleaner log here:


I was able to run it through the Analysis Tool to generate the following SNR graphs, too

All Sats

allsignalssnr.png

L1 Sats

L1SNR.png

L5 Sats

L5SNR.png

Sean Barbeau

unread,
Jan 12, 2021, 1:16:11 PM1/12/21
to GPSTest
Andrew,
So that looks much better for L5, right? It seems the "Force full GNSS measurements" setting worked?

I see some cutouts, but those are likely naturally lost signals.

Sean

and...@navmatic.io

unread,
Jan 12, 2021, 1:48:18 PM1/12/21
to GPSTest
Yup, "Force full GNSS measurements" seemed to be the trick.
The clock signal stays continuous, too.

Thanks again for the tip

Sean Barbeau

unread,
Jan 25, 2021, 12:53:56 PM1/25/21
to GPSTest
Andrew,
I wrote a blog post on this here:

Thanks again for the images! Hopefully people find this useful.

Sean
Reply all
Reply to author
Forward
0 new messages