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

Re: AO16/LO19 Traknet Operational!

> I'm still trying to figure out how to receive 1200 PSK, but i've gone
> through and carefully commented the 1200 AFSK 'soundmodem' code for
> LINUX and am ready to make a similar mod in software.  However, my 1992
> ARRL Handbook disappeared and its recent replacement while providing
> good overviews of a wider range of topics, often no longer contains
> information necessary for implementation by those not already familiar
> with a topic.  *Sigh*  That may be appropriate given a marked decrease in
> homebrewing and an increase in 'appliance operators', but also tends to
> perpetuate the comparative lack of innovative work by amateurs.  For
> example, i have neither the interest (nor the money right now) to go out
> and purchase the relevant books on communication theory containing the
> information in previous handbooks.

A 1200-baud Pacsat uplink signal may be generated by combining the data 
to be sent with a 1200 Hz clock (synchronized with the 1200-baud data)
in an exclusive-OR gate.  In doing so, the XOR gate is acting like a
balanced modulator, combining the data with the clock (1200-Hz carrier).
The output of the XOR gate is then low-pass filtered, and applied to the
audio input of a narrow-band FM transmitter, and Manchester encoded FSK
is the result.

It is interesting to note that while the output of this circuit is, in
fact, Manchester encoded data, it has all the properties of BPSK with
a carrier frequency of 1200 Hz!  In other words, a BPSK signal with a
carrier equal to the bit rate of the transmitted data is exactly the
same as a Manchester encoded data stream.

The 1990 ARRL handbook describes this type of modulation as "Bi-Phase -
Level", or Manchester II, or "split-phase".  By definition, a 1 produces a
half-bit period 1 pulse followed by a half-bit 0 pulse during the symbol
period.  A 0 does just the opposite (the phase is reversed -- flipped
by 180 degrees).

I went through the headache of figuring this stuff out in the early
90s when I first became interested in decoding BPSK data from OSCAR
satellites.  My first experimental demodulator consisted of only a few
chips that essentially extracted the BPSK carrier using a squaring loop
(a full-wave rectifier followed by an LM567 PLL) to regenerate the BPSK
carrier (since BPSK is generated in a balanced modulator, a "BFO"
operating at the bit rate is required to properly demodulate it).
The PLL's output then fed an exclusive-OR gate (balanced demodulator)
that was also fed with the input signal after passing through a limiter.
The output of the demodulator (XOR gate) was then low-pass filtered,
squared up again to TTL levels through a second limiter, and applied
to the TNC.

To my amazement, the circuit worked, but only worked well for FO-20 since
the other Pacsats running BPSK at the time had some phase jitter, and
nothing in my simple circuit could cope with anything but a perfect
signal.  Still, the thrill of success was as great as anything you
read on AMSAT-BB about making a contact through an Easy-Sat.  :-)
(Personally, I think the folks who don't get their hands into the
circuitry and software involved in OSCAR communications are missing
at least 75% of the fun, thrills, and feelings of accomplishment.
It also helps to expand and perpetuate the amateur satellite program
perhaps more than anything else.)

A GREATLY improved version of my Pacsat Modem was published in later
years in several publications, including the AMSAT Journal.  Since I
had gone through considerable labor in trying to understand how BPSK
and Manchester encoding worked (and got "stung" several times by
incorrectly published information), I made an extra effort to explain
how it all works and what functions the various stages of my modem
performed.  Last month I decided to convert the article to HTML and post
it on the Web.  Of course I couldn't find a text version the article,
so I had to resort to a BPSK test tape I had created for modem testing 
at the ARRL that contained the full text of the article sent as <UI>
frames.  I played the tape through the modem and recovered the text in
full.  :-)

Once I see my way clear (I'm still recovering from a hard disk failure),
I'll add the schematics and finish it up.  

73, de John, KD2BD

-=-=-=-=-=-=-=-=-=-=-=-=- John A. Magliacane, KD2BD -=-=-=-=-=-=-=-=-=-=-=-=-
Internet  : kd2bd@amsat.org          |  Voice : +1.732.224.2948
Satellite : AO-16, KO-25             |  Morse : -.-  -..  ..---  -...  -..
Packet    : KD2BD @ N2SMV.NJ.USA.NA  |  WWW   : http://www.njin.net/~magliaco/
Video     : 426.250 MHz/439.250 MHz  |  FAX   : +1.732.224.2060
-=-=-=-=-=-=-=-=-=-=-=- Linux Doesn't Cost.  It Pays. -=-=-=-=-=-=-=-=-=-=-=-

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