GPS Constellation Management: Playing not to lose

GPS Constellation Management: Playing not to lose
http://www.gpsworld.com/gps/gps-constellation-management-playing-not-l=
ose-9062-0?print=3D1

October 22, 2009
By: Eric Gakstatter

In sports, there is a phenomenon that sometimes occurs when a team is lea=
ding towards the=20
end of a game. It=E2=80=99s called =E2=80=9Cplaying not to lose=E2=80=9D.=


For example, there=E2=80=99s five minutes left in a basketball game and a=
team is leading by ten=20
points. The leading team wants to run the clock down as much as possible =
and still=20
maintain their lead. There are two basic strategies the leading team can =
take. One is to=20
continue being aggressive and using the strategy that put them in a winni=
ng position to=20
begin with. The other strategy is to try to =E2=80=9Cplay it safe=E2=80=9D=
until the time clock expires.=20
The problem with the latter strategy is that the other team can sense the=
change in=20
mentality and feed off of it. It=E2=80=99s called =E2=80=9Cplaying not to=
lose=E2=80=9D instead of =E2=80=9Cplaying to=20
win=E2=80=9D. I=E2=80=99ve seen it happen over and over again in team spo=
rts and in business. Once an=20
organization has achieved a level of success, they lose the edge that bro=
ught them their=20
success.

The reason you are seeing high HDOP warnings from the NAVCEN and GPS =E2=80=
=9Cbrownouts=E2=80=9D during=20
the day when RTK (GPS-only) isn=E2=80=99t working is because the GPS sate=
llite constellation is=20
sub-optimal. The current design of the GPS constellation is not focused o=
n =E2=80=9Cplaying to=20
win=E2=80=9D, but rather =E2=80=9Cplaying not to lose=E2=80=9D.

Even the original GPS Program Manager, Dr. Brad Parkinson, has voiced his=
concern about=20
GPS brownouts and discussed possible solutions. You can read one of his p=
resentations=20
here. In 2006, noted GNSS consultant John W. Lavrakas published a GPS Wor=
ld article=20
entitled Managing the GPS Constellation for Today=E2=80=99s Needs discuss=
ing the disparity between=20
the professional user community needs and GPS constellation management.

Today, there are 28 operational satellites. There were 30, but PRN08 is o=
ffline for=20
maintenance and PRN24 was placed in active reserve after an =E2=80=9Cunus=
ual failure=E2=80=9D. The 31st=20
one, PRN01/SVN49, never has been declared operational since its launch la=
st March due to=20
the issues discussed here before. The current GPS ground control infrastr=
ucture can only=20
handle 30 or 31 satellites.

Given these limitations, GPS looks grim for the GPS-only RTK user, right?=


Not necessarily.

The GPS constellation is optimized for 24 satellites. When there are more=
than 24=20
satellites in orbit, like there has been for many years, the extras are n=
ot positioned to=20
benefit the users but rather to be in a position to replace satellite fai=
lures. They are=20
sometimes referred to as =E2=80=9Cpaired orbits=E2=80=9D. Simply put, the=
active spares are orbiting very=20
near other satellites that are most likely to fail. This does very little=
for the user=20
community.

The current discussion is not whether to launch more satellites, but rath=
er how to=20
reconfigure the satellites that are in orbit. Launching more satellites i=
s a complicated=20
issue. It=E2=80=99s not just an US Air Force (the GPS stewards) technical=
issue, but a political=20
one because it=E2=80=99s expensive (~$150M per GPS satellite launch). Tha=
t leaves the Air Force=20
with the option of adjusting the GPS constellation to benefit the user co=
mmunity. Doing=20
this is not completely void of political implications I=E2=80=99m sure, b=
ut certainly not near the=20
risk of launching a new satellite and certainly a better bang for your bu=
ck to the user=20
community.

See the rest of the article, and the links implied above, at:
http://www.gpsworld.com/gps/gps-constellation-management-playing-not-l=
ose-9062-0?print=3D1

ot-...
ding towards the
is leading by ten
and still
take. One is to
ng position to
time clock expires.
change in
tead of =93playing to
usiness. Once an
ught them their
rownouts=94 during
constellation is
n =93playing to
concern about
resentations
ld article
e disparity between
ffline for
ilure=94. The 31st
st March due to
ucture can only
than 24
ot positioned to
lures. They are
es are orbiting very
for the user
er how to
s a complicated
, but a political
es the Air Force
mmunity. Doing
tainly not near the
ck to the user
ot-...

NAVSTAR was and is a system conceived to support the warfighters.
System reliability /availability is more important in that role than
absolute accuracy. An accuracy degradation from three meters to four
meters doesn't make much difference when delivering high explosive
ordnance. That's why the satellites are being clustered to back up
the key orbital slots.

I'd bet if you could get the system operators to offer an 'off the
record' comment, they would say we civil users have become a major
distraction.

--- CHAS

wrote:

http://www.gpsworld.com/gps/gps-constellation-management-playing-not-lose-9062-0?print=1
towards the
leading by ten
still
One is to
position to
expires.
change in
“playing to
Once an
them their
“brownouts” during
constellation is
“playing to
concern about
presentations
article
disparity between
for
The 31st
March due to
can only
24
positioned to
They are
orbiting very
the user
to
complicated
political
Air Force
community. Doing
not near the
the user
http://www.gpsworld.com/gps/gps-constellation-management-playing-not-lose-9062-0?print=1
I am a long time GPS user, I go back to a period when there were only
about 14 sats in the constellation. My experience was that by the time
there were 18 sats up, coverage world wide was 24/7 (it certainly
wan't with only 14). Sometimes the PDOP's were very bad (2 digit), so
while positional accuracy suffered (sometimes considerably), the
system was still usable. So as long as there are 24 operable birds
(and we have considerably more than that available today, and will for
the forseeable future), 2/7 coverage with reasonable PDOP is just
about a given.