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


I think that you are correct. New protocols need to support multipoint
links. This is probably one of the reasons that PSK31 is popular. Anyone can
scan around the working frequencies, listen to a few QSOs and contact an
interesting station. The other reason is that PSK31 works with extemely
modest stations because it works with a much lower signal-to-noise ratio
than RTTY and therefore needs less power. Nets are also impossible with
point-to-point protocols and nets are useful for building and maintaining
interest in a new mode. Other than voice communications, the other mode that
has garnered support is SSTV. Any new high-speed digital mode needs to
support streaming data and be capable of transporting voice, images and
text. We need something like AMTOR mode B that is fast enough to transport a
JPEG image in 10 seconds or less.

One problem with current PACSAT digital modes is that they require lots of
power because binary FSK without error correction requires a 15 dB
signal-to-noise ratio (SNR) to work reliably. PSK can work at a lower SNR
but is hard to use because of doppler shift and cannot be received reliably
at an unattended station.

I've been thinking about this off-and-on over the past year and one solution
is multitone FSK with Hadamard encoding of each byte. Multitone FSK spreads
the signals over a wider bandwidth than two-tone binary FSK but allows
reliable communication with a few dB SNR. It also tolerates the widely
variable group delay of the filters in amateur radio equiment. Since DSP
chips are now less than $20 each in small quantities and many PCs can now do
DSP internally, this method is cheap and hundreds of tones can be used. More
tones spread the information over a wider bandwidth but allow a lower SNR.
Given the limited power available in amateur satellites and the wide
bandwidths available in the UHF and microwave bands, this is a good
tradeoff. Multiple tones also have a high peak to average power ratio just
like SSB. This is a problem for one station, but may be an advantage when
multiple stations share a linear tranponder.

Hadamard encoding of each byte provides error correction if at least half of
the tones are received correctly. Use of 32 tones allows the transmission of
64 symbols for upper-case only ASCII (IA5) text. 128 tones supports 256
symbols for transmission of 8-bit bytes with complete transparency. One or
more extra tones allow recovery of the symbol clock at low SNRs and a
reference for doppler tracking.

At low speeds (appx. 10 bytes/sec.) this should allow the use of
omidirectional antennas for PSK31-like communication or APRS over P3D. At
high speeds, with directional antennas, image transmission is possible. Low
speed communication would fit within SSB bandwidths and work with existing
radios. Image transmission would require more bandwidth and not be
compatible with existing NBFM radios, but would require new radios or an
external adapter just as the new 38.4 and 76.8 kb/s pacsats do.

Is there interest out there in a new digital mode?



>Date: Tue, 5 Sep 2000 19:59:21 -0400 (EDT)
>From: Bob Bruninga <bruninga@nadn.navy.mil>
>Subject: [amsat-bb] HIGH SPEED Digital AMSAT Channels.
>BACKGROUND: 10 years ago, the PACSAT protocol optimized the delivery of
>data to all users in the footprint and allowed everyone to participate in
>the downlink.  It was GREAT.  It served the mode and all users well...
>NEW SITUATION:  Since then, and with the evolution of the Internet, there
>are now hundreds of DIGITAL applications that we will all like to
>experiment with through P3D and other new digital satellites.
>OBVIOUS SOLUTION?  At first glance, the obvious solutions are to take the
>Internet paradigm and apply it to our Ham satellites.  We gain
>tremendously from off-the-shelf hardware and software.
>PITFALL:  THis is all well and good for using our HAM satellites for
>point-to-point links, but it fails miserably at doing what ONLY HAM
>Satellites do best, and that is let everyone in on the action.  I fully
>support pt-to-pt links, but I also believe that the future of HAM radio is
>to RELISH our differences, not just do what we can already do on the
>Hams want to "tune the band", see what others are doing;  See what others
>are seeing...  Join in on a QSO or a "video conference".  Most of the
>straight-forward off-the-shelf things we may be thinking of doing via high
>speed Amateur channels just wont work except for full duplex between two
>stations A and B.  It will not let anyone else see or join the process
>easily without planning for it in the protocol.  We better be
>thinking about these things from the beginning...
>I fully support all aspects of experimentation.  But not until today, did
>it dawn on me how pt-to-pt links between stations A and B are of zero
>value to the other 1000 stations on the downlink who might want to be
>sharing in the action.  So, this message is NOT "against" anything (no
>flames please), it just suggests that coming up with protocols that allow
>everyone to play HAM radio (ie, read-the mail and tune in to a net in
>progress) is not going to be the simple off-the-shelf "internet" solution.
>Remember, there is NOTHING like HAM radio, and nothing like 1000 people
>trying to join in on the same downlink when most of them might be happy
>just participating in what someone else is doing.  Thus we need to be
>looking at how we can use one-to-many and many-to-many protocols...
>Just a thoght that came clear to me today...
>Again, nothing wrong with pt-to-pt.  We just need GROUP protocols too...
>de WB4APR, Bob

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