IP over DVB-S2

  • Hello,

    1) A kind of PACSAT BBS or HamNet access

    as described above...

    About the point 1, a short term solution for IP/Hamnet over QO-100, what do you think about my proposal to port my NPR protocol to QO-100 WB, like described in my previous post? Do you find that it would be useful or not?


    2) A successor for P4-A which would replace the Analog Linear Transponder Downlink by a regenerative digital transponder with a single DVB-S2 downlink.

    About the point 2, it is similar to what P4B, in the USA, planned to do a few years ago.

    In my opinion, it is far from optimal (it's only my opinion):

    - A pure "bend pipe", like QO-100, is much better for experimenters like us, in order for evey motivated OM to experiment new things on his own.

    - The number of simultaneous TX stations is (very) limited.


    Part of the Uplink would be still "linear" as it is for SSB/CW/etc., but just like a remote SDR which will stream the "passband" down as I/Q data on the DVB-S2 downlink.

    Another topic about "point 2" I’ve not understood the "narrow band SDR" that you propose, inside the satellite. If you want to capture 100kHz RF bandwidth at uplink side, then you need more than 24bits x 100kHz= 2.4Mbps for IQ samples of each channel, for 12bits IQ sample resolution, to the downstream. This induces a huge RF bandwidth need at downlink (~2MHz with QPSK) compared to a pure “bend pipe” solution (~100kHz). Therefore, unless I have not understood something, I don’t think that it’s a good idea.


    73, Guillaume F4HDK

    • Official Post

    Another topic about "point 2" I’ve not understood the "narrow band SDR" that you propose, inside the satellite. If you want to capture 100kHz RF bandwidth at uplink side, then you need more than 24bits x 100kHz= 2.4Mbps for IQ samples of each channel, for 12bits IQ sample resolution, to the downstream. This induces a huge RF bandwidth need at downlink (~2MHz with QPSK) compared to a pure “bend pipe” solution (~100kHz). Therefore, unless I have not understood something, I don’t think that it’s a good idea.

    Hi Guillaume,


    you are absolutely right! :/ I did this long ago and did not re-check, only circulated internally where nobody noticed this mistake either :huh: Maybe because we were also thinking in parallel about "virtual channels" with demodulation and regeneration, for example with 2.5 kHz BW and demodulator for SSB, PSK, etc. FM-Uplink, other modes with more BW indeed, etc.. Onboard demodulation would indeed save a lot of bandwidth.


    However, the idea to use a DVB-S2X only Downlink was lately driven by the fact that it would have a lot of advantage over an analog "bent-pipe" transponder. Mainly no need for drift compensation and abuse of uplink power would not have any effect. But it indeed also has disadvantages..


    On the other hand, a "STELLA" kind of Linear Transponder would probably even make more sense for SSB, CW, etc. This concept originally came from Howard Long G6LVB for P3-E and we are currently designing some new Hardware for a LEO satellite.


    But thinking more about it, it would in fact fit very well for a GEO...


    So perhaps a "bent-pipe" transponder with plug-in STELLA functionality and a separate DVB-S2X Downlink, but with DVB-S2 Uplinks and "regeneration"...


    I still need to learn more about NPR, it's protocol etc..


    There is also some development going by our team member Mario Lorenz DL5MLO regarding extremely narrow band DVB-S2X via QO.100, He was able to successfully receive the upper PSK FEC telemetry from the beacon just with an LNB directed to the satellite. So I hope this development can also give additional input..


    STELLA: Satellite Transponder with Equalizing Level Limiting Adapter




    73s Peter

  • This THREAD


    is quiet since some month.... (i dont know why) ....


    With DVB-S2 it is possible to send IP DATA to the full footprint....


    i tested my system with 120cm RX/TX and i saw 3Mbit/s in fullduplex mode uplink/downlink....with my own signal (DVB-S2 Mode VIDEO)


    Maybe IP over DVB-2 is not interesting for radio amateurs? (more for the commercial ones?)


    an idea to get some more for an RX .... with low money system can be found maybe here....


    https://www.heise.de/news/Def-…ter-300-Euro-4865615.html


    Edit: maybe there it is any possibble to get the ip stream also from the MiniTiouner RX? maybe also as interface for other applications like ...plutosdr for TX or plutosdr as fullduplex data transmitter???


    an IDEA also for MAKRO92???


    73 de dg0opk

  • Happy to see any discussion in this topic.
    I've got one of these for testing - this is DVB-S2 IP router. Only problem it works with 900Mhz IF minimum.
    It has linux onboard. Hacking into was needed because it doesn't have "factory reset" button.
    https://www.ebay.com/itm/Inter…-DVB-S2-DV3-/254219286566

    Our main problem is that now DATV is very complicated for "normal" hams. And with IP we will complicate this even further.
    What we need IP for ?
    We will build normal IP network and what ?
    I thought about sending internet in emergency situations. We could easily build 2Mb link...

  • Hi,

    I recently decided to experiment a bit with IP over DVB-S2 using Generic Stream Encapsulation. I ended up integrating libgse into leandvb and got some promising results, but the latency is high. I suspect this has something to do with my setup, but I did not extensively explore the problem.


    Essentially the application creates a virtual tunnel (TUN) interface that one can then manage as a normal network interface in Linux. Packets sent to that interface are received by the application and encapsulated using libgse and sent over the DVB link. GSE frames received from the link are de-encapsulated and written to the TUN interface.

    The way I implemented it requires you to always feed in a Transport Stream, even if you only intend to transmit GSE frames. It offers configuration options that allow you to split the bandwidth between GSE and TS data, and that ratio is fixed, meaning even if you do not have any IP packets coming to the interface (GSE traffic) the bandwidth allocated to GSE will not be used for TS data.

    I have had no time since then to make more experiments, but decided to post here in case someone is looking for something like this and/or is interested in playing with it.


    https://github.com/vankxr/lean…cd9398902b6428a0405194be7


    Regards,

    JS