Path: utzoo!attcan!uunet!husc6!cmcl2!nrl-cmf!ames!killer!netsys!cucstud!c3pe!c3engr!charles
From: charles@c3engr.UUCP (Charles Green)
Newsgroups: comp.dcom.modems
Subject: Re: 9600 baud over high-quality audio channel
Summary: Schmitt trigger recovers edges from ac coupling
Keywords: Schmitt trigger
Message-ID: <2856@c3engr.UUCP>
Date: 27 Jun 88 04:49:37 GMT
References: <17507@glacier.STANFORD.EDU> <5063@videovax.Tek.COM> <189@cbw1.UUCP>
Reply-To: charles@c3engr.UUCP (Charles Green)
Organization: C3, Inc. Engineering Division, Herndon, VA 22070-5201
Lines: 37

In article <189@cbw1.UUCP> brian@cbw1.UMD.EDU (Brian Cuthie) writes:
>In article <5063@videovax.Tek.COM> bill@videovax.Tek.COM (William K. McFadden) writes:
>>This might seem too obvious, but why not connect the output of your RS-232
>>port directly to the input of your audio channel (using suitable attenuation
>>and DC blocking)?  At the receive end you would probably want some sort of
>>active filtering and schmitt triggering to recover the data.  The problems I

>The first problem I see with this is that the audio path is not DC coupled
>and thus you would have big problems with data that was all marking.  Now if

I would expect his Schmitt trigger to take care of that; data that was all
marking would be seen as approaching 0 volts at the receiving end.  The trigger
will fire only on the edges of the input waveform, as the output voltage
exceeds the hysteresis.  For example:

       +V ---   ---                          ---   ---   ---
Input  0V              (data all marking)
       -V    ---   --------------------------   ---   ---   ---
                                             _
      ++V                                     --   _
       +V -__   -__  (ac coupling goes to 0)        --   -__      __
Output 0V             ____-------------------    __                 Threshold
       +V    _--   _--                          -     _--   _--   --

Note that after the marking condition, you have a higher-than-normal positive
voltage, but provided that your receive threshold is correctly set, you
should still be okay, although noise certainly needs to be dealt with.

>you MFM encoded the data you might get by, but I doubt it.  This is why
>people invented modems for phone lines (audio channels -- right ?).

I would expect that the technique of locking onto a frequency-shift signal
(a la Bell modems) to be relatively immune to noise, as is FM ("No Static at
All!").
-- 
#ident	"@(#).signature	1.5 :~charles/s..signature 04/27/88 21:40:43"
TwoBitNet: 460 Herndon Pkwy, Herndon, VA 22091-1928	UUCP:decuac!c3*!charles