Sarah and Walt, FYI the below email on Skynet communication. Upper
limits in uK would be valuable.
The Jan B2 tests (I've reattached the report) imply we should expect
pickup in a handful of channels at the ~100 uK level, and in median
channels at 10 uK, assuming 30 dB suppression due to being 40 deg out on
the Skynet sidelobe.
Sarah mentioned on the telecon today that 500 uK upper limit, though she
didn't say whether this was for every individual channel. That would at
least be consistent with expectations.
One could presumably take an average of RGL's to see if we can place an
upper limit on the mean level of pickup...
John
-------- Original Message --------
Subject: Re: skynet test [ping]
Date: Thu, 02 Aug 2012 16:05:13 -0400
From: John Kovac <jmkovac(a)cfa.harvard.edu>
To: Smith, Patrick D. <pdsmith(a)nsf.gov>
CC: John Ruhl <ruhl(a)case.edu>du>, John Carlstrom <jc(a)kicp.uchicago.edu>du>,
"Papitashvili, Vladimir O." <vpapita(a)nsf.gov>
Thank you Pat--we are also very interested in the SPTR-2 transmit times
John Ruhl requested.
In our own BICEP2 and Keck data from the brief NATO-4B 3.8 Hz test, we
are still working on determining whether features near 3.8 Hz which we
see in a subset of detectors are unique to the period of Skynet
transmission. In either case we will quantify the amplitude (or upper
limit) for 3.8 Hz in science data units (uK), and will compare to
expectations based on the January BICEP2 pickup test, extrapolated to 40
deg AZ offset--this should give us two ways to estimate contamination
levels and ultimately answer your question "how many dB suppression is
needed".
cheers,
John
On 8/2/12 2:46 PM, Smith, Patrick D. wrote:
John:
First things first. I just sent out the data call email, with you and
the others on this email on the cc list. Once we get that out of the
way, we can then think about coordinating a test.
We've been having equipment problems with the SPTR-2 RF chain, so I'm
not sure what our relative risks would be to perturb a currently working
configuration - not wanting to put at risk out operational service.
Also, I'm not sure what the technical considerations will be to
replicate at 3.8 Hz modulation (I'm assuming you are talking about a
pulsed amplitude modulation - 50% duty cycle square-wave of an
un-modulated RF carrier, on/off, at 3.8 Hz). We'd have to see how much
of the system configuration would have to be tinkered with to do this,
and then do a risk assessment to make sure we don't do something that
could put a return to normal ops at risk.
The default carrier modulation routinely used is a quadrature phase
shift keyed (QPSK) waveform at a 123 MHz data rate with Rate 1/2 forward
error correction (FEC). Here is an example calculation for the
transmitted carrier:
*Variable Name*
*Variable*
*Value*
*Units*
Data Rate
DR
123
MHz
Modulation Factor
m
2
QPSK
Viterbi FEC Rate
CRv
0.5
Rate 1/2
ReedSolomon FEC Rate
CRrs
0.921568627
Rate 188/204
*Symbol Rate*
*SR*
*133.4680851*
**
SR = DR / (m * CRv * CRrs)
QPSK has a sin(x)/x type of power spectra.
Pat
-----Original Message-----
From: John Ruhl [mailto:ruhl@case.edu]
Sent: Thursday, August 02, 2012 1:48 PM
To: Smith, Patrick D.
Cc: John Kovac; John Carlstrom; Papitashvili, Vladimir O.
Subject: Re: skynet test [ping]
Pat -
I'm not sure if I'm understanding you correctly… obviously when
talking to the satellite you have to use their spec'd modulation. Can
you do a TDRSS broadcast test (100% amplitude modulated at something
like 3.8Hz), not during a satellite window? You could be pointed a few
degrees away from the satellite, even, or do it when the satellite is
below the horizon.
thanks,
John
CWRU office: (216)368-4049
On Aug 2, 2012, at 1:43 PM, "Smith, Patrick D." <pdsmith(a)nsf.gov
<mailto:pdsmith@nsf.gov>> wrote:
John:
I'll see what I can do.
Re the modulation, due to operational concerns,
we're stuck with what
the system has to operate with. I believe that we are
using Rate 1/2
FEC QPSK modulation. We have to be compatible with NASA White Sands
requirements. The data encoding is using a CCSDS standard, so I could
ask if a spec on this could also be provided to you.
Pat
-----Original Message-----
From: John Ruhl [mailto:ruhl@case.edu]
Sent: Thursday, August 02, 2012 1:38 PM
To: John Ruhl
Cc: Smith, Patrick D.; John Kovac; John
Carlstrom; Papitashvili,
Vladimir O.
Subject: Re: skynet test [ping]
Pat -
By the way, it would be great to have a list of
transmission times
in both pointings, so we can compare.
thanks,
John
CWRU office: (216)368-4049
On Aug 2, 2012, at 1:33 PM, John Ruhl
<ruhl(a)case.edu
<mailto:ruhl@case.edu>> wrote:
> Pat -
> I would love to get your list of SPTR-2
transmission times; for the
last couple/few weeks would be a good start. We are
seeing some
interesting phenomenology in some of our detectors, but I don't know
whether RF pickup is a good model for that. It would also be helpful to
know what the broadcast signal looks like, ie carrier frequency and
modulation type.
>
> The skynet transmission was 100% modulated,
so much easier to look
for. Would it be possible to do a similar test with
SPTR-2?
>
> We are working hard to get the answer to the
question that you
asked, regarding suppression. We'll have much better
information on
that soon.
>
> thanks,
> -John
>
>
>
> CWRU office: (216)368-4049
>
>
> On Aug 2, 2012, at 12:43 PM, "Smith,
Patrick D." <pdsmith(a)nsf.gov
<mailto:pdsmith@nsf.gov>> wrote:
>
>> Thanks John, I appreciate the
information. Based on this
advanced-look information, the answer to my "how
many dB of suppression
do we have to do?" question is critical.
>>
>> Do you mind if I share your information
below with the Skynet team?
>>
>> Also, and I shudder to bring this up but
need to for your benefit
--- Have you seen any artifacts in your data that might
correlate to our
SPTR-2 earth station transmissions? This would be a harder phenomenon
to detect, as most of our SPTR-2 TDRS operations are with the satellite
located at 169.5 deg East, which puts Dark Sector clearly on the SPTR-2
antenna backlobe. However, on occasion, we use the TDRS satellite
located at 49 deg W, which means we have about a 48 deg angular
separation from Mapo from the SPTR-2 boresight. SPTR-2 uses a slightly
larger antenna (3.5 m v. 2.4 for Skynet), implying a narrower beam
pattern and better sidelobe supression, but one of the operating
frequencies (S-Band) is lower than Skynet (X-Band), which would imply a
broadening of the beam pattern. The uplink power levels will also be
different, but if you say you are seeing a high S/N in the NATO-4B
uplink bogey, I suspect that 3 dB or so in difference in SPTR-2 v.
Skynet terminal uplink power would not matter. I raise this because
you may be getting data contamination and not even know it. We can
probably get you a historical schedule of our SPTR-2 transmissions
(scheduled, at least) for any operations to the satellite at 49W. Are
you interested, and if so, how many days back would you want to go,
starting from today's date?
>>
>> Pat
>>
>>
>> -----Original Message-----
>> From: John Ruhl [mailto:ruhl@case.edu]
>> Sent: Wednesday, August 01, 2012 7:29 PM
>> To: Smith, Patrick D.
>> Cc: John Kovac; John Carlstrom;
Papitashvili, Vladimir O.
>> Subject: Re: skynet test [ping]
>>
>>
>> Pat -
>>
>> What I can report right now is that we
see the 3.8Hz modulated
skynet transmission with high S/N in our 150GHz detectors.
We are now
working to understand both what the likely mechanism for the pickup is,
and how this translates to science impact.
>>
>> We still have some fairly straightforward
analyses to do that
should help our understanding, but after that is done I think
we'll be
ready to have a discussion with you about paths forward.
>>
>> thanks,
>> John
>>
>> CWRU office: (216)368-4049
>>
>>
>> On Aug 1, 2012, at 4:40 PM, Smith,
Patrick D. wrote:
>>
>>> John R. and John K.:
>>> This is your semi-weekly ping. How
are things going?
>>>
>>> The Skynet team has just gotten word
from Intelsat that we can
make a long term swap in service from Skynet-4C to
NATO-4B, which should
give us the time to work the earth terminal relocation issue and
establish a new (hopefully long term) location that does not look over
Dark Sector. In the interim, we can continue service from the current
earth station location looking at NATO-4B at 34E. We are also going to
explore testing at 3 Mb/s, to see if we can operate at double the link
rate we had spec'ed.
>>>
>>> The last "i" to dot for
this provisional working CONOPS is the
outcome of your test data analysis.
>>>
>>> Eagerly awaiting the outcome.
>>>
>>> Pat
>>>
>>> -----Original Message-----
>>> From: Smith, Patrick D.
>>> Sent: Friday, July 27, 2012 8:42 AM
>>> To: John Ruhl
>>> Cc: John Kovac; John Carlstrom;
Papitashvili, Vladimir O.; Smith,
Patrick D.
>>> Subject: RE: skynet test
>>>
>>> Thanks John.
>>>
>>> The team is eagerly awaiting your
results.
>>>
>>> At issue for the team are the
following:
>>>
>>> 1. Ability to go into operational
communications on NATO-4B during
the remainder of this year and continue until ~
Feb 2014, by which time
we hope to have relocated the earth terminal to a better location with
less chance of interference. Going operational will help us realize a
return on investment of the ~ $650K spent on the space
segment/international backhaul for the year.
>>>
>>> 2. Decision on the site location to
move the earth terminal. We
should be doing pad/site work this coming austral
summer in preparation
for an installation in the 2013/2014 season, with commissioning by end
of Jan 2014. We still don't know if our proposed "Site 8" will work.
>>>
>>> 3. If Site 8 does not work in regards
to interference, then our
option is to try to push it even further north so that
Dark Sector is
behind (i.e., on the backlobe), and work on redesigning the antenna to
reduce backlobe energy spillage (e.g., tapered feed illumination,
reflector skirt, etc.). To move further to the north will invoke a
feasibility assessment to see if additional 4160 VAC power taps can be
created in the station to allow us the transmission power voltage we
need to extend the distance. As it is, the station cannot support any
additional 4160 power feeds, requiring an engineering effort to
address. This all adds cost, complexity, and time to the schedule. I
have to get all of this scoped and briefed to my management to see if
they are willing to fund it.
>>>
>>> 4. If we can't make a variant of
Site 8 work at all, our only
recourse is our "Site 10", which is in the
backlobe of SuperDARN's HF
radar emissions, and we aren't sure what the EMI effects would be. We
would have to get busy with some engineering to try to
estimate/anticipate the interference issues to see if it is worth the
risk to move to Site 10.
>>>
>>> For reference, updated site selection
drawings are attached so you
can refresh your understanding of where these two
locations are.
>>>
>>> Therefore, the results of your data
analysis of the NATO-4B test,
plus a comment on my question re "how many dB of
suppression do we need
to supply to push us into your noise", are in the critical path for
resolving the long term deployment of the earth station, and time is
getting short to converge on a final site to enable initial site prep
work to occur this coming season.
>>>
>>> No pressure, or anything.... :-)
>>>
>>> Cheers,
>>> Pat
>>>
>>>
>>> -----Original Message-----
>>> From: John Ruhl
[mailto:ruhl@case.edu]
>>> Sent: Friday, July 27, 2012 6:27 AM
>>> To: Smith, Patrick D.
>>> Cc: John Kovac; John Carlstrom;
Papitashvili, Vladimir O.
>>> Subject: Re: skynet test
>>>
>>> Pat -
>>>
>>> Sorry to not have gotten back to you
before your meeting.
>>> We're seeing some
"interesting" behavior, still sorting it out but
making good progress.
Reliable results soon, I hope!
>>>
>>> -John
>>>
>>> CWRU office: (216)368-4049
>>>
>>>
>>> On Jul 25, 2012, at 9:10 AM, Smith,
Patrick D. wrote:
>>>
>>>> John^2:
>>>>
>>>> How is the data analysis coming?
I have a Skynet project team
meeting coming up tomorrow, and I am sure to be pinged
on status. I'd
like to give the team an update on when we can expect to get briefed on
the results of the latest test re NATO-4B.
>>>>
>>>> FYI, in case you haven't
heard yet, the USAP Blue Ribbon Panel
Report has been publically released. There
is a chapter on
Communications and IT, and there is a significant reference to South
Pole broadband communications and the need for improvement.
>>>> See:
>>>> i
>>>> ca
>>>> _
>>>> 07232012.pdf (pp. 129-139)
>>>>
>>>> Rgds,
>>>> Pat
>>>>
>>>>
>>>> -----Original Message-----
>>>> From: John Kovac
[mailto:jmkovac@cfa.harvard.edu]
>>>> Sent: Wednesday, July 18, 2012
7:02 AM
>>>> To: John Ruhl
>>>> Cc: Smith, Patrick D.; John
Carlstrom; Papitashvili, Vladimir O.
>>>> Subject: Re: skynet test
>>>>
>>>>
>>>> Hi Pat,
>>>>
>>>> I second that--these are
excellent questions and we're working on
the answers. Our analysis team is
hard at it this week--we'll be
comparing our findings with John Ruhl and discussing them asap.
>>>>
>>>> John
>>>>
>>>>
>>>> On 7/17/12 11:19 PM, John Ruhl
wrote:
>>>>> Pat -
>>>>>
>>>>> All good questions. Until we
get our analysis straight I don't
think we can answer any of them, but
we'll get there!
>>>>>
>>>>> As soon as we and the
bicep/spud crew have final numbers from
this test, we'll talk.
>>>>>
>>>>> thanks,
>>>>> John
>>>>>
>>>>> CWRU office: (216)368-4049
>>>>>
>>>>>
>>>>> On Jul 17, 2012, at 7:40 PM,
Smith, Patrick D. wrote:
>>>>>
>>>>>> John:
>>>>>>
>>>>>> Thanks for the update. I
will be anxious to see the results of
your analysis.
>>>>>>
>>>>>> To the best of my
knowledge, the protocol at Pole in place is
to record all on/off times and to
ensure that the personnel at Pole know
what is going on. I will verify just to make sure.
>>>>>>
>>>>>> A couple of questions:
>>>>>>
>>>>>> 1. What do you think it
will take to track down the the 3.8 Hz
artifacts that you are seeing to trace the
source? It seems that the
test we've run is inconclusive if we find this ghost but can
definitively determine its origins.
>>>>>>
>>>>>> 2. Is there any way you
could guess as to how many dB of
suppression you would need in our signal to make
it drop into your noise
background, given a good known case of detection, such as was for the
tests conducted with Skynet-4C this past austral summer? Perhaps I'm
just being a dumb engineer at this point, but it seems to me that if we
have a verified interference event coupled to a known point on the
antenna beam pattern and a known power level into the antenna feed, if
you could deduce from the data how much suppression would be needed to
make it drop into the noise, we could then have a hard engineering
performance target for our future design efforts to relocate the earth
station to eliminate the interference.
>>>>>>
>>>>>> I'm a bit in the
dark regarding the possible observing modes
>>>>>> where long term
integration of data from the same spatial
>>>>>> location would render the
above simplistic approach not workable.
>>>>>> From prior emails, I get
the sense that this is a real concern,
>>>>>> and one that is hard to
quantify in a reasonable time period.
>>>>>> We have a real tough
problem to solve in that the logistical
>>>>>> umbilical effect that
prevents us from building a new Skynet
>>>>>> terminal too far away
from the main station also has Dark Sector
>>>>>> close-in. If we are
going to devise an engineering solution
>>>>>> (say, use a bigger
antenna and/or a tapered illumination, use of
>>>>>> an offset feed; use RF
absorbing foam to attenuate side/back
>>>>>> lobes, etc.), we will
need some sort of attenuation suppression
>>>>>> value to design to. If
you're derivations indicate that we
>>>>>> need>100 dB worth of
suppression, then we know that will be
>>>>>> unattainable and we have
a major problem. I'm just trying to
>>>>>> figure out if there is
any logical way to get a handle on this
>>>> to figure out a solution without
taking years to sort this out.
Any ideas would be most welcome.
>>>>>>
>>>>>> Pat
>>>>>>
>>>>>>
________________________________________
>>>>>> From: John Ruhl
[ruhl(a)case.edu]
>>>>>> Sent: Monday, July 16,
2012 4:22 PM
>>>>>> To: Smith, Patrick D.
>>>>>> Cc: John Kovac; John
Carlstrom; Papitashvili, Vladimir O.
>>>>>> Subject: Re: skynet test
>>>>>>
>>>>>> Pat -
>>>>>>
>>>>>> I just wanted to update
you on our analysis... we are working
hard on it. We see some weak peaks at 3.8Hz
in some of our channels,
but it's not clear yet whether that's due to skynet, whether we have the
normalization correct yet, and therefore what our estimate of potential
contamination from skynet would be.
>>>>>>
>>>>>> We will keep you posted.
In the meantime we understand that
you need to keep skynet as a backup. Please do
let us know if/when
skynet transmits again; so far our winterovers have informed us of two
transmissions, one that was a mistake and a second that was for a long
test of connectivity.
>>>>>>
>>>>>> thanks,
>>>>>> John
>>>>>>
>>>>>>
>>>>>> CWRU office:
(216)368-4049
>>>>>>
>>>>>>
>>>>>> On Jun 25, 2012, at 6:38
PM, Smith, Patrick D. wrote:
>>>>>>
>>>>>>> John -
>>>>>>> I'm working via
Vladimir as the overall lead for working this
>>>>>>> issue with the
astronomy community. I'm handling the
>>>>>>> contractor interfaces
at Pole and on the design/support team
back here in the US.
>>>>>>>
>>>>>>> Points for your
consideration:
>>>>>>>
>>>>>>> We've made a
major physical reconfiguration change of the
>>>>>>> pointing of the
Skynet terminal to point to NATO-4B's
longitude at 34E.
>>>>>>> The terminal only has
a +/- 10 deg longitude adjust capability.
>>>>>>> The physical reconfig
change puts the functioning of the
>>>>>>> terminal at risk due
to concern of damage, and we only
implemented this due to the
>>>>>>> confirmed
unsuitability of working via Skynet-4C at 1W this
winter. I
>>>>>>> will not want to put
the terminal at risk of further damage to
>>>>>>> reconfig to point at
Dark Sector, and then undo to repoint back
>>>>>>> to 34E, if at all
possible.
>>>>>>>
>>>>>>> I do have
instructions from Brian Stone to ensure that the
>>>>>>> Skynet terminal can
be brought on-line and made operational as
>>>>>>> an operational
contingency in the event of a prolonged outage
>>>>>>> of the
>>>>>>> GOES-3 system that
puts operations and/or life-safety at the
>>>>>>> Pole at risk. At
this point in time, based on what we know, it
>>>>>>> would appear much
less harmful to use NATO-4B as this
contingency backup in lieu of Skynet-4C.
>>>>>>>
>>>>>>> Our desire is to go
operational on NATO-4B this winter, but we
>>>>>>> won't do so
without running the proper evaluation with the
>>>>>>> astronomy community.
It will be Vladimir's call. However, we
>>>>>>> need to be mindful of
my management direction regarding
emergency back-up capability.
>>>>>>>
>>>>>>> In the big picture
sense, we do need to find a way to make
this work.
>>>>>>> GOES-3 has only about
3 years of life left in it until we need
>>>>>>> to start seriously
looking at disposal to prevent an orbit
>>>>>>> debris hazard. For
NSF to let GOES-3 get stranded in the
>>>>>>> gravity well at 105W
will be a black eye to the Foundation and
>>>>>>> will be greeted with
disapproval from the international space
operations community.
>>>>>>> Given this, we need
to implement a GOES-3 follow-on.
>>>>>>> Skynet-4C, with
possible use of NATO-4B until we can sort out a
>>>>>>> suitable "Grid
North" earth terminal location to hopefully
>>>>>>> solve the
interference issue with Dark Sector, is the only
alternative that we have at
present.
>>>>>>>
>>>>>>> Intelsat and
Paradigm/Astrium have been very supportive in
>>>>>>> working with us to
try to implement the NATO-4B work-around and
>>>>>>> let us substitute it
for Skynet-4C until we sort out our
>>>>>>> permanent earth
station. I don't want to unduly prevail on
their good graces if we can
possibly avoid it.
>>>>>>> I understand your
team's commitments for the SPIE conference.
>>>>>>> I just hope that
there aren't any further delays.
>>>>>>>
>>>>>>> Best Regards,
>>>>>>> Pat
>>>>>>>
>>>>>>>
>>>>>>> -----Original
Message-----
>>>>>>> From: John Ruhl
[mailto:ruhl@case.edu]
>>>>>>> Sent: Monday, June
25, 2012 12:41 PM
>>>>>>> To: Smith, Patrick
D.; Papitashvili, Vladimir O.
>>>>>>> Cc: John Kovac; John
Carlstrom
>>>>>>> Subject: skynet test
>>>>>>>
>>>>>>> Pat and Vladimir -
>>>>>>>
>>>>>>> I'm writing to
update you on the skynet testing; I'm not sure
>>>>>>> who our actual POC is
on this.
>>>>>>>
>>>>>>> As you probably know
the winterovers ran the test last week.
>>>>>>> Unfortunately some of
the key people on our analysis teams
>>>>>>> (John Kovac says the
same is true for them) are headed to an
>>>>>>> SPIE conference soon
and I expect that we will not be able to
>>>>>>> fully analyze things
until they return in a couple weeks.
>>>>>>>
>>>>>>> We'd like to
reiterate that even if we do not detect the signal
>>>>>>> in this roughly 2
hour test, that is not a guarantee that it
>>>>>>> won't show up in
our season-long maps. After analyzing this
>>>>>>> test, we'd like
to step back and think about what other testing
>>>>>>> or procedures we
should put in place, given the limited ability
>>>>>>> to change the test
conditions (ie repoint the skynet antenna
>>>>>>> significantly) during
the winter. There has been some email
>>>>>>> traffic that we saw
via the winter overs indicating that people
>>>>>>> were thinking about
starting regular operations with skynet. I
>>>>>>> hope we can put that
off until after we've all had a chance to
>>>>>>> talk once we've
done our final analysis. If tranmissions are
or have been made, we want to know
the start and stop times, of course.
>>>>>>>
>>>>>>> thanks,
>>>>>>> John
>>>>>>>
>>>>>>> CWRU office:
(216)368-4049
>>>>>>>
>>>>>>>
>>>>>>
>>>>>
>>>>
>>>>
>>>> --
>>>>
___________________________________________________________________
>>>> John Kovac
jmkovac(a)cfa.harvard.edu <mailto:jmkovac@cfa.harvard.edu>
>>>>
>>>> Assistant Professor, Astronomy
and Physics, Harvard University
>>>> 160 Concord Ave rm 310, Cambridge
MA 02138, 617-496-0611
>>>
>>
>
--
___________________________________________________________________
John Kovac jmkovac(a)cfa.harvard.edu
Assistant Professor, Astronomy and Physics, Harvard University
160 Concord Ave rm 310, Cambridge MA 02138, 617-496-0611