Subject: Lacrosse: What's up with the antenna...?
Date: 28 Jan 1997 10:11:51 GMT
From: aufsj@imap2.asu.edu
Newsgroups: sci.space.policy
Organization: Arizona State University
Lines: 98
Message-ID: <5ckjd7$j7i@news.asu.edu>

	The two antennas on the FAS Lacrosse representation (assuming it 
is accurate) may, as posted previously, be signs of a spinner.  For more 
"normal" spacecraft using GPS for attitude determination 4 antennas is 
much more common ( "Gadfly" comes to mind ), and if simple input into 
post ephemeris generation is wanted, one will usually do.  As I said, on a 
top-dollar top-flight spacecraft one would expect everything to be done 
for a reason, and why would one include two antennas? (two receivers in 
case of breakdown is more understandable, but fixed antennas rarely break 
and it is relatively simple to connect two redundant receivers to a 
common antenna).  In addition, the attitude data that such a GPS set-up 
could provide could be crucial if the bird is indeed SIGINT (perhaps more 
later on that).

	But what *really* keeps me from feeling at ease with this 
spacecraft ID is the antenna.  The antenna is where "the rubber meets the 
road," and as much admiration as I have for the engineers who don black I 
doubt that they have pulled off any miracles of science.  Why is there a 
paucity of imaging radar satellites that use dish antennas?  As posted 
earlier, the Magellan probe was an exception because it had to dual use 
its radar antenna with communications, and because as a mapper it could 
afford to simply wander all over the planet.  Presumably, US intel types 
are not going to want to wait 45 days or whatever to get a picture of 
what they want, so the radar signal has to scan.
	Why a big dish?  They are typically used to increase the "gain" 
of the collected signal, so weaker things can be picked up. They also 
usually are more selective about where they pick up from.  They tend to 
be very sensitive in the direction they are pointing, decidedly 
unsensitive in other directions. In effect, the dish "points" a radar 
signal at some point in space (or listens most sensitively in the same 
direction, if it is not emitting).  
	So the antenna is only 'looking' at a small area. If the antenna 
is fixed, then, you would only be able to image some small strip of land 
corresponding to the ground track of your orbit. This means that 
"revisit" times would be very large, like Magellan, and this does not 
look very useful.  One has to scan the radar signal.  Normally, this is 
done by using an array as an antenna.  
	If one were to use a dish, however, what then?  Well, you could 
mechanically scan the dish, just like you can with a typical ground based 
radar.  There are some problems, however.  For starters, the motors and 
servos and whatnot that it takes to do this are a real danger, moving 
parts and spacecraft do not always mix well.  In addition, getting a 
large mass to move in relation to another while in zero-G and in orbit is 
challenging.  For example, as related in many accounts from the 70s, US 
geo SIGINT satellites had to be "boresighted", i.e. their antennas were 
fixed and the whole spacecraft had to be moved to point at a new place.  
Apparently, "just moving" the antenna to a new aim point is harder than 
it may seem.  In addition, the FAS rendition shows a somewhat slender and 
uncomplicated attachment between the large antenna and the main body of 
the spacecraft. If the antenna were going to be moved about wouldn't one 
expect more control mechanisms and joints and such?
	The other alternative is to move the whole payload in order to 
point the antenna.  In other words, the spaceraft could presumably "roll" 
or "spin" and still be used to collect radar imagery.  However, the data 
from the Belgian Working Group tends to support the idea that the 
platform is either not spinning, or it is spinning very fast--in which 
case it would probably be useless for radar imagery.  It is possible, of 
course, that the thing is rotating along some axis at a reasonable speed 
but that the paint job or reflections or whatever just aren't showing it 
off.  Guess I'll have to dig my 10meter telescope out of the closet ;-).
	So the possibilities seem to be:  It is spinning really fast, and 
it has just escaped notice.  It is not spinning.  It is spinning but the 
collective "we" can't tell.  Well, that narrows things down a bit!

	But this brings up the matter of stabilization. The bird on the 
FAS page doesn't appear to have any spin stabilizing components. This 
would tend to lead to one of two conclusions at first glance: The whole 
platform is spun (antenna and all), or the thing is 3-Axis stabilized 
with most probably a set of gyros (There don't appear to be any gravity 
gradient booms).  The gyros would presumably take up some the 'box' space 
on the main platform. But this once again leads us to a big question: If 
the thing is stabilized, how do they aim the goldarned antenna?  Of 
course, if it is spinning at a rate reasonable for imagery collection 
then why hasn't it been noticed?  Or is it spinning fast, and not doing 
imagery at all (that dang option keeps sneakin' in).
	This weaves back in the issue of GPS.  If the bird is collecting 
imagery then precision ephemeris data (particularly with regards to 
rotational equilibrium) is not that important.  The locations of imaged 
objects could presumably be corrected by scene matching, i.e. Here is a 
mountain we know is at point X, this blip is Y away==now we can correct 
for bias and error.  A single GPS antenna could provide input data for 
general ephemeris creation.
	If, however, the bird is passively collecting data, then 
extremely precise spacecraft attitude data becomes vital---knowing the 
main antennas pointing angles at all times can be crucial.  And, as I 
have posted before, it is my understanding (be advised: I'm not an expert 
by any stretch) that a two-antenna GPS configuration is really only 
useful if the parameters are changing very quickly, as with a spinning 
spacecraft.
	I'll bring up frequency and heritage issues in a new post.

regards,

----------------------------------------------------------------------
Steven J Forsberg  at  aufsj@imap2.asu.edu              Wizard 87-01