g4klx / DAPNETGateway

Gateway to the DAPNET POCSAG network.
GNU General Public License v2.0
32 stars 22 forks source link

Problems decoding transmit sequences with two or more messages... #35

Open dl9sec opened 4 months ago

dl9sec commented 4 months ago

Hi Jonathan,

first of all, thank you very much for your effort with the DATNETGateway :-)

Since a few days I have the DAPNETGateway activated at my Pi-Star MMDVM Dual Hat Hotspot and had the opportunity to delve into it quite intensively and try out a few things. While waiting for my Alphapoc 602R I used my Flipper Zero together with the "Pocsag Pager" app (and also a universal Sub-GHz analyzer app) to see what happens on-air at my DATNETGateway.

An very interesting fact was, that the Flipper had big problems to decode transmit sequences from my hotspot which contained two or more messages (they were shown in the Pi-Star tableau with the same timestamp). Whenever such a sequence apppeared, the Pocsag decoder mixed up the messages and gave the RIC from the last received message. Then I got my Alphapoc and observed, that in case of these "multi-message" transmit sequences the Alphapoc decoded and displayed only the first message of the sequence and discarded the rest until the next hotspot transmission.

I described this also at the Pi-Star forum: https://forum.pistar.uk/viewtopic.php?t=4921

For me it seems, that the gap between those sequenced messages within one transmission is too short and therefore the decoders doesn't get all messages correct and so they are lost.

Would it be possible to have the gap time between messages transmitted in one transmit cycle configurable and maybe as a parameter in the Pi-Star expert configuration? Or at least to get a experimental version of the DAPNETGateway, where the gap time is longer, to see if the decoding gets better?

Thank you in advance.

Best regards, Thorsten DL9SEC

EA4GFC commented 4 months ago

Como curiosidad a mi también se me pierden mensajes si os sirve de algo saludos 73

EA4GFC

El mar, 7 may 2024 18:47, Thorsten Godau @.***> escribió:

Hi Jonathan,

first of all, thank you very much for your effort with the DATNETGateway :-)

Since a few days I have the DAPNETGateway activated at my Pi-Star MMDVM Dual Hat Hotspot and had the opportunity to delve into it quite intensively and try out a few things. While waiting for my Alphapoc 602R I used my Flipper Zero together with the "Pocsag Pager" app (and also a universal Sub-GHz analyzer app) to see what happens on-air at my DATNETGateway.

An very interesting fact was, that the Flipper had big problems to decode transmit sequences from my hotspot which contained two or more messages (they were shown in the Pi-Star tableau with the same timestamp). Whenever such a sequence apppeared, the Pocsag decoder mixed up the messages and gave the RIC from the last received message. Then I got my Alphapoc and observed, that in case of these "multi-message" transmit sequences the Alphapoc decoded and displayed only the first message of the sequence and discarded the rest until the next hotspot transmission.

I described this also at the Pi-Star forum: https://forum.pistar.uk/viewtopic.php?t=4921

For me it seems, that the gap between those sequenced messages within one transmission is too short and therefore the decoders doesn't get all messages correct and so they are lost.

Would it be possible to have the gap time between messages transmitted in one transmit cycle configurable and maybe as a parameter in the Pi-Star expert configuration? Or at least to get a experimental version of the DAPNETGateway, where the gap time is longer, to see if the decoding gets better?

Thank you in advance.

Best regards, Thorsten DL9SEC

— Reply to this email directly, view it on GitHub https://github.com/g4klx/DAPNETGateway/issues/35, or unsubscribe https://github.com/notifications/unsubscribe-auth/APQM2RYKDUMZDUTAYS2CLQDZBEATDAVCNFSM6AAAAABHLLZMTWVHI2DSMVQWIX3LMV43ASLTON2WKOZSGI4DGOBRGAYDOOI . You are receiving this because you are subscribed to this thread.Message ID: @.***>

dl9sec commented 4 months ago

EA4GFC

Tnx for the support :-)

It is very obvious that messages are lost if one subscribes to transmitter group "dxclusters" and e.g. "DX KW" (RIC 1004). In the evening there are a lot of messages which are transmitted in a sequence. If comparing to the Pi-Star tableau and what is received by the pager you can see that messages are lost (and this is not because of RF signal loss)...

73, Thorsten