[Ground-station] Experimental Channels

[Wally Ritchie]

On Wed, Feb 12, 2020 at 3:13 PM Paul Williamson via Ground-Station
<ground-station at lists.openresearch.institute> wrote:

> Allow me to expound on a few of Wally's points:
>
> > Using such channels will require allocating an experimental IQ channel
> > through the DMT's standard protocols available to the most basic of
> > stations.
>
> Yes. One of the key benefits of the all-digital DMT design is that uplink
> users are subject to automatic authentication (proof of identity) and
> authorization (permission to use the system) before they are permitted to
> communicate through the system. Any pseudo-analog IQ relay channel would
> have to be set up on demand by request of an authenticated, authorized
> user, for a limited period of time. The requesting user would then become
> responsible for the channel, sort of like the control operator on a
> terrestrial repeater. In particular, the responsible user would be expected
> to monitor the content of the downlink channel created, and shut it down if
> abuse is detected.
>
> That scheme means that Phil's proposal to accommodate FM uplinks for users
> so inclined is much less interesting than it might first appear. The
> hypothetical FM users would have to also operate a digital uplink to set
> the channel up. It wouldn't replicate the experience of a plain analog FM
> transponder. Of course, even without the auth+auth scheme, there's no way a
> microwave GEO can replicate the key advantage of FM LEOs: the use of cheap
> off-the-shelf VHF/UHF FM transceivers and low-gain hand-waved antennas. The
> users already have to obtain specialized equipment to operate the DMT, and
> we are already planning to do our best to cost-optimize it; they might as
> well use it in the intended all-digital way and get all the advantages
> designed into the system.
>
> The experimental IQ relay channels are not about satisfying users who
> think they want an old-fashioned "analog experience". It's about
> experimenting with alternative uplink waveforms that the payload is not
> programmed to demodulate and multiplex onto the downlink. Presumably the
> goal would be to eventually implement the waveform in the payload,
> eliminating the gross inefficiency of downlinking raw IQ data. With a
> reprogrammable payload, that might be possible on the same mission.
> Depending on the details, it might only be possible in a follow-on mission
> with more powerful hardware on board.
>
> > Such general purpose IQ facilities can also provide familiar waterfall
> > displays where the power spectral density of a band segment can be
> > transported over the downlink for a or 2D/3D waterfall display.
>
> They could, but that's not the right way to do it. A waterfall display
> needs only a tiny fraction of the bit rate an IQ relay would require for a
> given bandwidth. The payload should do the work and the rate limiting.
>
> > In practice, the spectral density display of the uplink will likely be
> > an always available feature available in the downlink stream.
>
> I'm not so sure about that. Someone with more of the link budget figures
> at their fingertips will have to check me on this, but I'm guessing that
> normal uplink signals at appropriate power levels should be pretty hard to
> see on a waterfall. A full-bandwidth waterfall would only be useful for
> diagnosing loud interfering signals. Such a tool would not need to be
> always on. It might even encourage miscreants to transmit loud signals just
> to show up. The payload managers would need access to a waterfall
> occasionally, but it might be counterproductive to train users to watch it
> routinely.
>
Of course it depends on the specific implementation of the multi-channel
receiver(s) but in many cases there will be a front-end fft on the order
2*11 or so dividing a channel into multiple slots prior to overlap and add
functions. So in these cases there can be sort of free psd's that can be
tapped and used also for waterfall etc. As an example, even 2048 over 2 MHz
is about 1K per bucket. This is still useful for channels even as small as
5K. It can be useful as a visualization aid to see these and how they
behave over time. For most likely distributions the energy distribution is
pretty even for a given signal except at the start and end. But the
different channels will probably see 10 db or so power differences. For
analog experimentation, the dynamic range might be much higher but for
experimentation with analog, a lot could be changed on the fly. In fact,
it's not too far fetech to be a sort of digital spectrum analyzer
accessible to anyone on the ground.

I guess the whole point here that everybody agrees on is that even though
we have fully digital downlink, it's providing quasi error free data and we
can use that to support a lot of analog analysis, even though it may be
inefficient. Those doing research would probably be using 8PSK, 8/9 modes
on the downlink to transport the streams they are looking at which will be
about 5 times more spectrally efficient than a baseline QPSK 1/4. It's a
great use of wideband capability even though it might be looking at narrow
analog signals. The possibilities on the input side are endless. All the
more reason to get working DVB-S2 downlinks into space ASAP.

WU1Y


>   -Paul KB5MU
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openresearch.institute/pipermail/ground-station-openresearch.institute/attachments/20200212/70265889/attachment.html>