iMet5 not decoding

[HB9EVW]

Hi,

does anybody know why iMet5 signals are not being decoded by my Raspbery Pi 5 and auto_rx 1.7.4?

I can clearly hear 2 very strong signals on my Yaesu VX-6R radio, and both are visible on the auto_rx scan plot, but neither can be decoded! Auto_rx tries to, but ends up with error.

The signals are fluctuating in strength (simiiar to any RS41 I have heard), so they are probably flying or moving.

Any idea what could this be?

[hb9tmc]

I receive them too, 402.000 and 404.000 with QSB. Nobody seems to decode them.
For me they are currently too weak for a useful recording. Maybe you can record it for later analysis?

73
Stefan

[HB9EVW]

Hi Stefan,

Thanks for the reply. Exactly, they are on 402 and 404 MHz. and are quite strong here! I will try to record them with my SDR receiver. I recorded the audio already, but will try to record the signal itself.

Vy 73,

Ayar

[HB9EVW]

So, here is a baseband recording of the 404 MHz signal made with an RTL-SDR. The signal is strong enough but had some fading at the start!

The file is about 500mb.

https://drive.google.com/file/d/1WTlE4iEhdv6lcBXasI19jg1oIYLd3n7h/view?usp=sharing

vy 73

[HB9EVW]

Here is an audio of the signal as well...

[hb9tmc]

The signal strength increased a bit here until it vanished. I got at least some plausible data out of it from the imet54 decoder, I've attached it.
I'll keep the audio recording in case someone is interested for further analysis.

73
Stefan

[HB9EVW]

Great! Thanks for the decoded data! So, it is iMet5 after all! Of course iMet5 normally uses AFSK modulation, so I know that my ttgo receivers cannot decode their signals as it only uses FSK, but auto_rx should be able to decode it as far as I know!

But anyway, it should be possible to decode the data from the audio alone.

A question: what is the imet54 decoder you have used? 

vy 73.

[hb9tmc]

Yes, it seems to have been another experiment from Payerne.

I was using the decoder from autorx (./auto_rx/imet54mod)
But at the moment I don't know why the output wasn't used. Mark or someone else can surely tell.

[HB9EVW]

I see, so it is part of auto_rx. Well, we have to wait and see what Mark says.

Let's hope that this iMet5 sonde was only an experiment, and not something Payerne will use on a regular bases! RS41 is probably still the cheapest and its launch is also fully automated, but who knows!

Thanks for the info and have a nice evening!

Vy 73

[hb9tmc]

My best guess is that they are testing R2Home with different types of sondes. Since it was two of them.
But unlike the RS41, they had both about the same signal strength. I have only the recording from the signal on 404.0 MHz.

[HB9EVW]

Yeah, it might have been an R2Home test. At the beginning, (at least at my QTH near Bern), both signals were at the same level. At one point, the 402 became weaker, then disappeared altogether while the 404 was still strong until it was gone later . So this could indicate that the 402 was the R2H one. We anyway need to keep an eye on these tests! They have experimenting quite a lot lately!

[Mark Jessop]

Yes, this looks to be an iMet-5x (could be an iMet-50 or an iMet-54).

Unsure if there's enough SNR to decode. The signal is 4800 baud FSK (not AFSK!).

Cheers,

Mark

--
You received this message because you are subscribed to the Google Groups "radiosonde_auto_rx" group.
To unsubscribe from this group and stop receiving emails from it, send an email to radiosonde_auto...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/radiosonde_auto_rx/8247326c-f689-43fd-8f0e-42a24b60cc03n%40googlegroups.com.

[Mark Jessop]

I've merged in rs1729's changes into the auto_rx testing branch (v1.7.5-beta3). If you update to this hopefully these sondes will decode!

73

Mark VK5QI

[HB9EVW]

I had to delete my previous message because Vigor is not the authour of imet54mod as I wrongly wrote, it is rs1729 of course who is the author. Sorry for the confusion. The main thing is that it works now!