[Date Prev][Date Next][Thread Prev][Thread Next] - [Date Index][Thread Index][Author Index]

Re: Amateur radio module on commercial Geo Sat?


On Wednesday 20 December 2000 06:18, Phil Karn wrote:
> >Just one wee little question.  How would TDMA work, when there's stations
> >spread across a wide area (with vastly different, and generally large
> >propagation delays), without being either horrendously slow, or stations
> >accidentally falling out of their timeslots, due to propagation delays?
> >That's one of the limitations of GSM mobile phones in rural areas here,
> >incidentally.
> You are indeed correct that this is a serious problem with TDMA in
> terrestrial cellular with overlapping cells and omni mobile
> antennas. That's one of the reasons Qualcomm invented IS-95 CDMA.

In fact GSM's ideas (I assume we are talking about GSM here) is not so bad.
A frame (about 125 us in length) consists of 8 timeslots per carrier. A BTS which
occupies only one channel one of those timeslots contains the so-called BCCH
(Broadcast Control CHannel). This channel is a kind of beacon that provides frequency
and symbol synchronization to the users. Moreover it supplies status and provider
The mobil phones are basically in standby mode (just monitoring the paging slots) and
become active (i.e. transmit) only when loging into the network, during operation and
when changing the so called location area. A location area consists of a number of base
stations strategically grouped together. For example, base stations serving a highway can
be grouped together in one location area over some 50 miles in order to reduce the number
of location area updates transmitted by clients who are driving along the road.
However, if a mobile has to transmit is does not just do that at a random point of time when
the channel is not occupied from its POV as we do in our packet radio network (CSMA).
There are time slots which are designated "access slots" in uplink direction. The mobile
transmits a "access burst" which is about half the length of the slot length. The burst is so
short because at this point of time the distance to the base station is not known. If it is very
high and the burst was longer a collision with the subsequent time slot would occur.
We could basically do the same in our design. The other thing is, if two mobiles transmit
an access burst at the same time they collide. Both back up and try again at a later
time according to a stochastic process. Finally, if an access burst reaches the
base station a temporary control slot is assigned and the time difference caused
by the propagation path is made known to the mobile (TADV = "Timing ADVance") so that
it can compensate for it. I think we could do the same, too.
By the way, in case of GSM the resolution is 550 metres and the maximum distance
(=maximum cell radius) is about 35 km, so I do not really understand what is meant by
talking about  "limitation of GSM mobiles in rural areas". At 35 km distance at least
a 1800 MHz signal is probably attenuated to much to enable sensible voice communication

Back to our problem. Of cause a similar scheme to that pointed out above would require
a significant amount of "intelligence" aboard the bird in question. However, this way
also misbehaved stations who fiddled with the protocol stack parameters could be locked
out from service quite trivially, too.

At last, do not forget that we have a considerable amount of processing power on the
rudak unit on AO-40. In case it can be recovered again I think we should make use
of this fact and not condemn the fine hardware to be expensive variants of 1200 bps AFSK
and 9600 bps FSK modems.

  -- Jens
Via the amsat-bb mailing list at AMSAT.ORG courtesy of AMSAT-NA.
To unsubscribe, send "unsubscribe amsat-bb" to Majordomo@amsat.org