phasing with respect to LES-9 and ATS-3 is essential to maintain necessary contact with
winter-overs and experiments. Next year the combination of LES-9 and GOES-3 will satisfy
a reasonably phased connection to the Internet. They may also satisfy data throughput
As Figure 1 indicates, F-3 and GOES-3 are visible simultaneously. So while F-3 may not be
necessary after November 1994, it should still be considered as an emergency backup.
Below is a list of requirements, deficiencies, and concerns we have been able to identify in
our discussions with many research groups at South Pole:
1. Continue to identify and acquire existing usable satellites for
a. Higher throughput and
b. Internet connectivity several times during the day
Assuming that launching our own satellites is out of the question, the most reasonable
approach to increase the communications capability at the Pole is to continue to identify
and acquire existing usable satellites, either free (LES-9, ATS-3, GOES-3) or leased
(MARISAT F series). Current rates for Marisat F3 are $15K/month for 1 hour/day at
9.6Kbps or $29K/month for 1 hour/day at 56Kbps. These costs translate to $145/MByte
and $48/MByte, respectively.
Table 1 shows that GOES-3 will take most of the throughput load, as well as being above
the horizon about 12 hours from LES-9. It is a single point failure and we cannot depend
on its services forever. Even if it operates trouble free, it is not clear what will happen
after the agreement with NOAA expires. While we may expect a rate between 256-512
Kbps, it is questionable whether we can achieve T1 rates. Tests by Paul Eden (no earlier
than August 95) will help determine maximum rates. If it is possible to achieve T1 rates,
purchases for modems and (perhaps) dishes may be required.
It is also likely that, in a few years, TDRSS will be available for ultra high data
bandwidths out of the Pole. We should pursue this possibility.
2. Better connection dependability
LES-9 can one day be very good, and another day be absolutely terrible. This is
apparently caused by fading and ionospheric conditions. While it is unlikely that we can
improve the performance, there are cases in which poor performance can be improved
(bringing LES 9 from 24 Kbps up to 38.4 Kbps by changing modulation techniques is an
example). A plan for ensuring satellite reliability should be developed and thoroughly
In addition, the sparing and documentation for these critical satellite systems appear to be
minimal at best. Time should be spent documenting what now exists.
3. Phone service to CONUS from the work areas
Currently, the ATS-3 phone link requires the winter-over to make the phone call from the
Comms building. Not only is this inconvenient, but impractical. It is very important to
make things as convenient and productive as possible. We recommend developing
procedures to allow calls to be made at any phone on station. This should become the
standard, continuing when GOES-3 and Marisat are available.