Path: utzoo!utgpu!watserv1!watmath!att!tsdiag!davet
From: davet@tsdiag.ccur.com (Dave Tiller N2KAU)
Newsgroups: comp.sys.ibm.pc.hardware
Subject: Re: Baud vs. bps.
Message-ID: <1084@tsdiag.ccur.com>
Date: 29 Aug 90 13:44:34 GMT
References: <849@idcapd.idca.tds.philips.nl>
Organization: Concurrent Computer Corp. Oceanport NJ
Lines: 46

In article <849@idcapd.idca.tds.philips.nl> lexw@idca.tds.PHILIPS.nl (Lex Wassenberg) writes:
>
>Okay, in some recent articles there were these hints about the difference
>between "baud" and "bits per second". Now I for one have always thought that
>baud and bps is the same. So what's the difference? Has it something to do
>with bits that don't carry actual data like start and stop bits? Or is it
>more subtle? Thanks to anyone who can inform me (and others).

A good question, Lee.  I'll give it a shot.  The term 'baud' refers to the
number of symbol transitions per second (note that the unit 'per second' is
built into the name baud.  Bauds/s is bogus).  Bps refers to the actual number
of data bits sent per second.  These two numbers can be the same, or the
bps figure can be greater than the baud rate.  How can this be? Well, there's
more than one way to modulate a signal, and there's more than one way to
represent data than 0's and 1's.  Here's an example:

If I choose to use phase modulation, I could easily use the following:

    Phase change      dibit
    ------------      -----
         0 deg         00
        90 deg         01
       180 deg         10
       270 deg         11

In this scheme, each phase transition (baud symbol) represents _two_ data
bits! Therefore, if I decide to sent 2400 transitions/sec (baud rate=2400)
the actual bps rate will be 4*2400=4800.  Neat, huh?  Taken to an extreme,
you can simultaneously change the phase and amplitude of a signal, stretching
the number of data bits/symbol higher and higher.  Another method to squeeze
data over a narrow bandwidth channel is to use multiple tones, instead of
just two for mark/space.  Telebit's PEP protocol starts out with _512_(!)
discrete tones and negotiates down from there depending on line quality.
If all the tones were used for data, at 19200 bps the actual baud rate would
be 37.5!!!  I'm pretty sure there's lots of error check/correction stuff as
well, so don't expect this is really the case.  On a related note, the
FCC was quite visionary when they declared that the baud rate of HF data
transmissions can't exceed 300 _baud_.  If we can squeeze more bits/baud
out of our signal, all the better. With some new DSP modems, I can forsee
4800-9600 bps over a 3 kHz voice channel to almost anywhere in the world.
I hope this explanation helps, mail me directly if you have any questions.
-- 
David E. Tiller         davet@tsdiag.ccur.com  | Concurrent Computer Corp.
FAX:  201-870-5952      Ph: (201) 870-4119 (w) | 2 Crescent Place, M/S 117
UUCP: ucbvax!rutgers!petsd!tsdiag!davet        | Oceanport NJ, 07757
ICBM: 40 16' 52" N      73 59' 00" W           | N2KAU @ NN2Z