[Ground-station] Potential Phase 4 - Gateway interference

[Wally Ritchie]

One of the features of DVB-S2 is that Gold sequence are used in the final
physical layer scrambler. These are derived from an 18 bit number n. The
use of Gold sequences has the effect of reducing the correlation between
two signals which reduces the effects of co-channel interference.  It can
also be used to identify particular DVB-S2 sources.

By default, broadcast applications use a default n = 0 so that receivers
can descramble all signals without knowing the value for n. But for
applications such as ours, n can be set to a different value for each
DVB-S2 source to minimize co-channel interference. The ASIC chips we are
looking at have this capability.

When DVB-S2 receivers are validated, the validation involves both the
wanted signal and one or more adjacent and/or co-channel unwanted signals
at various power levels (as well the usual AWGN and clock phase noise
impairments). This is an important component in the process of "validating"
a receiver for an application, either in simulation or on real hardware.
When we talk about the performance of a receiver it is not just the
performance over a clean AWGN channel that is relevant but rather the
performance in the RF environment of the final system. So this is always
the caveat as well as one of the reasons that additional margin is almost
always required beyond theoretical link budgets, a fact that many find easy
to ignore.

I agree that it doesn't appear likely that we would have much interference
between LOP-G and GEO for reasons pointed out by Kent. However, it's not
too early to think about some band-planning. We have frequency re-use for
geo slots but LOP-G will follow the moon. We have 50MHz allocation up and
down (if we can hold on to it). This is not all one would ever need any
more than 640K is all the memory one would ever need. (I'm writing this on
a machine that is +50dB over that and servers processing it are another
+10dB over that).

As an aside, we might also want to consider Gold codes for the uplink. It's
relatively easy at both ends and it's likely they could improve adjacent
channel interference at little cost.

WU1Y






On Fri, Feb 14, 2020 at 4:22 PM Michelle Thompson via Ground-Station
<ground-station at lists.openresearch.institute> wrote:

> OK I see what you mean.
>
> In doing a BOE for GEO, the occlusion is longer lasting, because the moon
> moves more slowly across the sky than a lower satellite, but the outcome
> appears to be the same. It's rare.
>
> If we can put this on a single slide, for that one time someone
> with authority requires an answer, then the work to come up with numbers is
> worth it. I think Kent is correct in the order of magnitude for LEO and
> pretty close numerically.
>
> I am interested in generating a solid number. Does anyone have good
> software for this, or want to write up a python script that encapsulates
> the problem?
>
> -Michelle W5NYV
>
>
>
> On Fri, Feb 14, 2020 at 1:07 PM KENT BRITAIN <wa5vjb at flash.net> wrote:
>
>> Interference would be rare, and momentary.
>>
>> OK, the bird goes though a 360 deg arc in about 80 minutes.
>>
>> So in a typical overhead pass that's about 4 deg a min..
>> ( Yes, a lot of variables in that equation! )
>>
>> The antenna beam is 6 deg if we allow for the -10 dB points.
>> So 90 seconds would put the 2nd bird well out of the pattern from the
>> ground station.
>>
>> Lots of orbits, a parallel orbit is not impossible, but in most cases they
>> would only be in common view of the ground dish for a minute or so.
>> I'll let someone else do this calculation.  Take two LEO Birds, and when
>> looking from a common point on the ground, how often do they pass within
>> 6 deg of each other?    Not often I suspect.    Kent WA5VJB
>>
>>
>>
>> On Friday, February 14, 2020, 1:17:53 PM CST, Michelle Thompson <
>> mountain.michelle at gmail.com> wrote:
>>
>>
>> Good point on frequency management.
>>
>> We plan to use (up to) the entire 10MHz amateur allocation.
>>
>> Gateway appears to be using less.
>>
>> Given that one can model this as a narrowband interferer to a broader
>> band signal, in your opinion, can we accept or ignore the potential
>> interference?
>>
>> -Michelle W5NYV
>>
>>
>>
>>
>> On Fri, Feb 14, 2020 at 10:20 AM KENT BRITAIN <wa5vjb at flash.net> wrote:
>>
>> The typical Direct TV style dish looks at 1/4000 th of the sky.
>> Well, we only see 1/2 of the sky at any given time, so even if both birds
>> are visible
>> at the same time, only 1/2000 of the time would they both be in the same
>> beam
>> even if on the same frequency.
>>
>> Yea, can crunch these numbers several ways but with a little frequency
>> management
>> interference is going to be pretty rare.    Kent WA5VJB
>>
>>
>>
>>
>> On Friday, February 14, 2020, 12:03:12 PM CST, Michelle Thompson via
>> Ground-Station <ground-station at lists.openresearch.institute> wrote:
>>
>>
>> With AREx and Lunar Orbiting Platform Gateway (usually just called
>> "Gateway") developing and progressing, Paul KB5MU brought up something we
>> need to check on.
>>
>> If Gateway uses the 10GHz amateur downlink (and they plan to) how much
>> impact do other 10GHz payloads present? QO-100, Phase 4X, CatSat, etc.?
>>
>> I think any interference or outages are going to be very small, but I'd
>> like to have an answer before someone like ARISS asks.
>>
>> If we all use DVB-S2/X, then it would seem that there would be a
>> potential for brief interference.
>>
>> Assuming 10GHz downlinks from Gateway
>> 1) how often would potential interference occur for GEO, HEO, LEO?
>> 2) is there anything that we (or Gateway) should do about it?
>> 3) is there anything that we (or Gateway) could do about it?
>>
>> Goal: Quantify and document.
>>
>> -Michelle W5NYV
>>
>>
>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openresearch.institute/pipermail/ground-station-openresearch.institute/attachments/20200215/05f903f3/attachment.html>