A229 Pro weirdness (timestamp skips, front/rear video sync, GPS data issues)

[NicholasOakenscowl]

(I haven't updated the firmware; it was already the V1.2_240227 when I got the camera)

Problem 1: Timestamp jumped backwards by minutes while driving
A couple of days ago, my A229 Pro did something weird during one clip (1min):
12:27:24 clip starts, while driving in an underground car park
12:27:47 surface reached, driving continues
12:28:09 everything's normal, no GPS yet...
12:24:13 wtf? The time just jumped backwards by about 4 minutes!
12:24:15 GPS coordinates appear on the video
12:24:27 clip ends

As such, going by filenames my clips were out of order:
20240612_122623_F
20240612_122723_F <- problem happened here
20240612_122427_F
20240612_122527_F

Fortunately the time didn't jump by exactly 1 minute, or it would have presumably overwritten a file!
What could be the cause of this?

The A119S added an incremental ID to each video file name, so the true order of the files was recoverable even when the time changed (time zone changed or similar).
I wish this was added to the A229 Pro as well.

Problem 2: Front and rear video/timestamps not in sync
Front and rear camera video/timestamps in their respective files are not in sync. For example, if I pick the first clips after startup:
2024_0608_171924_F: timestamp in frame 1 is 17:19:25 and it takes 17 frames to change to 17:19:26
2024_0608_171924_R: timestamp in frame 2 is 17:19:24 and it takes 53 frames to change to 17:19:26
(frame 2 on rear camera because frame 1 lacked the overlay).

This is quite annoying when vertically stacking front and rear footage. If the difference was constant, it wouldn't be so bad, but it varies. The most I've had to compensate for so far was 1.28 seconds!

Problem 3: GPS data quality
Sometimes the GPS data is completely nonsensical, such as after exiting an underground space. Looking at extracted GPX data (which matches the overlay GPS info), there's obviously something going on, as I evidently took a quick (well, supersonic) trip to Sweden:
<trkpt lat="64.182129" lon="19.450297"><time>2024-05-22T17:14:19Z</time><speed>329.702023</speed><course>352.720001</course></trkpt>
<trkpt lat="64.180591" lon="19.449821"><time>2024-05-22T17:14:20Z</time><speed>353.258402</speed><course>9.320000</course></trkpt>
<trkpt lat="64.179053" lon="19.449341"><time>2024-05-22T17:14:21Z</time><speed>332.438869</speed><course>352.510010</course></trkpt>
<trkpt lat="64.179053" lon="19.449341"><time>2024-05-22T17:14:21Z</time><speed>332.438869</speed><course>352.510010</course></trkpt>
After this, I returned to Finland.
<trkpt lat="60.175399" lon="24.807654"><time>2024-05-22T17:14:23Z</time><speed>9.990503</speed><course>252.880005</course></trkpt>
<trkpt lat="60.175431" lon="24.807463"><time>2024-05-22T17:14:24Z</time><speed>9.825881</speed><course>253.479996</course></trkpt>
<trkpt lat="60.175407" lon="24.807300"><time>2024-05-22T17:14:25Z</time><speed>9.301147</speed><course>253.210007</course></trkpt>
<trkpt lat="60.175391" lon="24.807137"><time>2024-05-22T17:14:26Z</time><speed>8.899881</speed><course>251.779999</course></trkpt>
<trkpt lat="60.175366" lon="24.806982"><time>2024-05-22T17:14:27Z</time><speed>8.709537</speed><course>254.410004</course></trkpt>

I like to extract the GPX data into Spatialite, and the A229 Pro sure does like to travel! Here's its GPS tracks over a little more than a month:

(do notice the short visit to the coast of Kolguyev Island)

My A119S would also occasionally glitch a bit, but in its case it was generally just 1 second having an erroneous longitude. Here's its GPS tracks over 6 years in several different cars:


This one may be partly my fault, though, since I've fitted the A229 Pro on the roof console rather than windshield glass. While this has the upside of keeping the camera mostly out of direct sunlight, and being perfectly in the middle, it does unsurprisingly delay establishing GPS fix.

 

[Nigel]

Sometimes the GPS data is completely nonsensical, such as after exiting an underground space. Looking at extracted GPX data (which matches the overlay GPS info), there's obviously something going on, as I evidently took a quick (well, supersonic) trip to Sweden:

Hmm, just a suggestion, but maybe your "sometimes" ties up with this map:

GPSJAM GPS/GNSS Interference Map

Map showing possible GPS interference around 61.257, 21.868 on 2024-05-22.

gpsjam.org

If it happens just after exiting an underground space, maybe it has locked on to a fake GPS signal as the only one it can see, until it finds a decent one with good signal quality? There are not going to be many strong GPS signals that far north at the best of times, so it may need to wait to find a Galileo satellite to get a good reading.

There have been some airline routes abandoned in that area recently due to jamming of the GPS signals:

Finnair suspends flights to Estonian airport due to ‘GPS interference in the area’

The approach to Tartu is dependent on GPS – and the Russian border is only 40km away

www.independent.co.uk

 

[JardaB]

The GPS module must have direct visibility to the satellites. The trees are a problem, the forest is a problem, the streets between the tall buildings are a problem.

It should see the whole sky from horizon to horizon to see more than as a bare minimum 4 satellites, optimally much more 8-16, at most around 20 satellites

 

[NicholasOakenscowl]

Hmm, just a suggestion, but maybe your "sometimes" ties up with this map:

http://[URL]https://gpsjam.org/?lat=61.25681&lon=21.86849&z=3.2&date=2024-05-22

[/URL]

If it happens just after exiting an underground space, maybe it has locked on to a fake GPS signal as the only one it can see, until it finds a decent one with good signal quality? There are not going to be many strong GPS signals that far north at the best of times, so it may need to wait to find a Galileo satellite to get a good reading.

There have been some airline routes abandoned in that area recently due to jamming of the GPS signals:

http://[URL]https://www.independent...russia-ukraine-finnair-safe-gps-b2536985.html

[/URL]

I had thought about this being related to the jamming, but that is an excellent point about it possibly locking on to a fake signal before getting a real fix!

The fake GPS signal would explain problems 1 and 3. The camera reacquires a fix very quickly if the car's been parked outside, or after exiting a tunnel, but actually parking undergrounf (home, work, shopping centres, etc) seems to be a problem.

The GPS module must have direct visibility to the satellites. The trees are a problem, the forest is a problem, the streets between the tall buildings are a problem.

It should see the whole sky from horizon to horizon to see more than as a bare minimum 4 satellites, optimally much more 8-16, at most around 20 satellites

Longer fix time aside it has actually generally worked quite well I think, aside from the occasional glitches. Still, I might consider moving it to the windshield instead, if I can find a good spot. The car's own "camera hump" is quite large, and fitting the camera right beside it results in said hump blocking the view a bit. Moving it away from the hump would make the camera stick out like a sore thumb, while moving it ahead of the hump would put it right in the driver's field of view.

I'd love to have a camera mount with an external GPS unit. Viofo, are you listening?

 

[Nigel]

It is quite possible that there is interference coming from your car cameras, which seem to be very close to the GPS receiver, but the communications should have error detection and correction, and I've not heard of other people having corrupt data, so I think the possibility of fake GPS transmissions should be taken seriously, if very expensive aircraft GPS has problems sufficient to ground an air route then there must be something pretty nasty being transmitted.

but actually parking underground (home, work, shopping centres, etc) seems to be a problem.

The GPS receiver is supposed to remember where all the satellites are located, so shouldn't have a problem with finding them after parking underground, unless it is for a period of over a week.

I'd love to have a camera mount with an external GPS unit. Viofo, are you listening?

Yes, in your case it looks like it would be useful, maybe Viofo could allow the GPS to be in the rear camera instead of the front camera?