Language/OS - Multiplatform Resource Library/LANGUAGE OS.iso/dsp/dspgroup/dspmails.txt

Internet Message Format | 1990-03-08 | 97.3 KB

From ka7axd  Tue Jul  4 14:18:55 1989
Received: by ka9q.bellcore.com (5.60/4.7)
    id AA22424; Tue, 4 Jul 89 14:18:55 EST
Date: Tue, 4 Jul 89 14:18:55 EST
From: ka7axd (Mike Horne)
Message-Id: <[e-mail]>
To: dsp-local
Subject: yat


Yet another test.

Mike

From [e-mail] Tue Jul  4 18:35:32 1989
Received: from ucsd.edu by ka9q.bellcore.com (5.60/4.7)
    id AA00156; Tue, 4 Jul 89 17:35:22 EST
Received: by ucsd.edu; id AA27103
    sendmail 5.60/UCSD-1.0
    Tue, 4 Jul 89 13:05:59 PDT
Received: from ka9q.bellcore.com by ucsd.edu; id AA27092
    sendmail 5.60/UCSD-1.0
    Tue, 4 Jul 89 13:05:41 PDT for /usr/lib/sendmail -odq -oi -ftcp-group-request tcp-group-list
Received: by ka9q.bellcore.com (5.60/4.7)
    id AA23545; Tue, 4 Jul 89 14:30:35 EST
Date: Tue, 4 Jul 89 14:30:35 EST
From: [e-mail] (Mike Horne)
Message-Id: <[e-mail]>
To: [e-mail]
Subject: dsp-group@ka7axd roll call (so far)...
Status: R


> Since Mike's having problems with a mailer on the other side of...

Contrary to comments from Brian, the [e-mail] is alive and
kicking (ka7axd is not experiencing any mailer problems, either side of
relay.cs.net).   Thank you for the volunteer help, though, Brian.  As of
today (in the wee hours of the 4th o' July) there have been several messages
regarding switched cap filters, the Ariel dsp board, the 56K, etc.

Currently the dsp-group is comprised of the following dsp-interested people:

[e-mail]                (Richard Bisbey - NG6Q)
[e-mail]                (Bob Albrightson - N7AKR)
[e-mail]            (Maitland Bottoms - AA4HS)
[e-mail]                    (Curtis Spangler - N6ECT)
[e-mail]            (Tom Clark - W3IWI)
[e-mail]                (Dave Borden - K8MMO)
[e-mail]            (Dave McNeill - WB7BLB)
[e-mail]            (DSP archive pseudo-person)
[e-mail]            (Bdale's redist list)
[e-mail]            (Phil's redist list)
gatech!stiatl![e-mail]        (John DeArmond - WD4OQC)
gatech!uflorida!novavax!ankh![e-mail]  (Garry's redist list)
[e-mail]                (Greg Troxel - N1DAM)
[e-mail]                (Glen Elmore - N6GN)
[e-mail]            (Glen R. Sands - KA7AYF)
idacrd![e-mail]            (Bob McGwier - N4HY)
[e-mail]            (Jim English - WA7QLB)
[e-mail]                (John A. Limpert - N3DMC)
[e-mail]            (Michael Horne - KA7AXD)
[e-mail]                (Jim Mankin - KB3KJ)
[e-mail]                (Dave Trulli - NN2Z)
[e-mail]                (Phil Lapsley - N6TCT)
[e-mail]                (Paul Shen)
uunet!ism780c!kodak![e-mail]    (Barry L. Ornitz - WA4VZQ)
uunet!nikhefh.hep.nl![e-mail]    (Henk Peek - PA0HZP)

Again, to get your very own subscription to the dsp-group, send your name,
callsign, and a valid email address to:

    [e-mail], or
    [e-mail], or
    ...{backbone}!uunet!ka7axd.wv.tek.com!dsp-group-request

To post messages to the group, send them to:

    [e-mail]

or appropriate permutation as shown above.  Brian's guidelines that he has
set forth for the tcp-group should also be followed here (re: redistribution,
etc.).

I maintain archives of the dsp-group on a per-month basis, or finer resolution
as dictated by the volume of traffic.  To receive archives, send the month/year
span of interest to [e-mail], and they will be
promptly shipped out to you.  Better yet, if you have Internet access, the
archive can also be accessed via anonymous ftp from:

    ka9q.bellcore.com (128.96.160.1)

in /pub/dsp/dsp-group.

Please send any questions or comments directly to me.  Enjoy!

Mike
[e-mail]


From @relay.cs.net,@tektronix.tek.com:[e-mail]  Tue Jul  4 19:13:12 1989
Received: from flash.bellcore.com by ka9q.bellcore.com (5.60/4.7)
    id AA00307; Tue, 4 Jul 89 19:13:12 EST
Received: from RELAY.CS.NET by flash.bellcore.com (5.58/1.1)
    id AA00148; Tue, 4 Jul 89 20:13:48 EDT
Received: from tektronix.tek.com by RELAY.CS.NET id aa07118; 4 Jul 89 20:14 EDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA29298; Tue, 4 Jul 89 17:15:31 PDT
Received: by orca.wv.tek.com (5.51/7.1)
    id AA00746; Tue, 4 Jul 89 17:11:14 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA02727; Tue, 4 Jul 89 17:15:43 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA02725; Tue, 4 Jul 89 17:15:40 PDT
Received: by orca.wv.tek.com (5.51/7.1)
    id AA00726; Tue, 4 Jul 89 17:07:49 PDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA29270; Tue, 4 Jul 89 17:11:53 PDT
Received: from RELAY.CS.NET by zephyr.ENS.TEK.COM (4.0/7.1)
    id AA29683; Tue, 4 Jul 89 17:09:18 PDT
Received: from ka9q.bellcore.com by RELAY.CS.NET id aa06501; 4 Jul 89 20:05 EDT
Received: by ka9q.bellcore.com (5.60/4.7)
    id AA00290; Tue, 4 Jul 89 19:03:03 EST
Date: Tue, 4 Jul 89 19:03:03 EST
From: Mike Horne <[e-mail]>
Message-Id: <[e-mail]>
To: [e-mail]
Subject: Re: DSP modems...
Cc: [e-mail]


> I want a DSP in my PC that can emulate the all the non highspeed modems
> (say < 20Kb) at the same time...
> I known that there is done some work on packetradio-DSPs. (TAPR and ???).

Well, Henk, you are correct.  I've been working on some Motorola 56K based
hardware for the PC, and apparently AMRAD is working on DSP-related projects
as well.  And of course, AMSAT/TAPR have been working on TMS32010-based
hardware/software; Bob or Tom are probably the best ones to get a recap of
their current work.  Bob has been working with the 56K recently, as well.
Anyone else care to crawl out of the closet? :)

There really isn't any reason why you couldn't implement various modems
on a single DSP processor board (though some processors have more horsepower
to do the job easier than others).  I think you can expect to see a drop-in
board for the PC Real Soon Now, and other stand-alone boards for the rest
of the computer world in the future.  Commercial drop-in boards already exist
for most of the DSP processors that are out today.

>What I really want to known is, is there a "standard" DSP and how many modem
>and hdlc uart and hdlc-driver combinations you can emulate at the same time on
>a DSP (board)?

There really isn't a `standard DSP' out there, but TI's 32010-series parts
have seen a lot of design-ins.  Quickly taking over the low-end DSP world is
the Motorola 56K-series (mainly the 56001).  It provides some impressive
power for a relatively low price (~$50 US for SLAM pack).  Probably following
these parts in performance are the TI 320C25 and C30.  The C30 is an entrant
into the world of floating point (FP) DSPs; the other parts are all
fixed-point/ integer devices.  There are other miscellaneous fixed-point
DSPs by NEC, Zoran, AT&T, etc.

There are many new FP parts out, or coming out, and their performance is
very impressive.  Motorola's 96K-series (mainly the 96002) will be a real
workhorse when it comes out later this year.  AT&T's DSP-32C is a very
interesting part, mostly because the assembly language looks very similar to
C.  One of the most impressive parts for dedicated DSP applications is the
new Intel i860, mostly because of its dual-instruction mode, dual execution
units (FP and integer core), and its sheer speed (effective 25 nS execution
time when massaged correctly).  It wasn't originally designed for DSP
applications, but its internal architecture lends itself nicely to these
types of problems.  It's main drawback for DSP applications is its price;
currently ~$800 US.  All of these FP parts (including some fixed-point
devices) have C compilers available for them, but for low-level, high
throughput routines you'll *have to* resort to assembly language.  Intel
and others are promising C compilers with intelligent pipeline schedulers,
but that remains to be seen.

RAM to feed these suckers is somewhat expensive at this time.  32KX8, 35 nSec
parts are running about $50 US, but I know of at least one vendor who will
be dropping the prices down to about $30 US for these parts in the middle of
this month.

As far as how many modem-types you can emulate, it is nearly limitless.  It's
strictly a function of how much horsepower you have, how easy it is to
convert the modem functionality into software, and how good of a hacker you
are. :)

> Henk, PA0HZP  ..!uunet!mcvax!nikhefk!henkp.UUCP

Mike
(followups to the [e-mail])



From @relay.cs.net,@tektronix.tek.com:[e-mail]  Wed Jul  5 14:18:01 1989
Received: from flash.bellcore.com by ka9q.bellcore.com (5.60/4.7)
    id AA01196; Wed, 5 Jul 89 14:18:01 EST
Received: from RELAY.CS.NET by flash.bellcore.com (5.58/1.1)
    id AA14446; Wed, 5 Jul 89 15:18:41 EDT
Received: from tektronix.tek.com by RELAY.CS.NET id aa25166; 5 Jul 89 15:16 EDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA20804; Wed, 5 Jul 89 12:17:19 PDT
Received: by orca.wv.tek.com (5.51/7.1)
    id AA22750; Wed, 5 Jul 89 12:12:25 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA04160; Wed, 5 Jul 89 12:17:14 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA04158; Wed, 5 Jul 89 12:17:07 PDT
Received: by orca.wv.tek.com (5.51/7.1)
    id AA22691; Wed, 5 Jul 89 12:08:51 PDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA20627; Wed, 5 Jul 89 12:12:49 PDT
Received: from RELAY.CS.NET by zephyr.ENS.TEK.COM (4.0/7.1)
    id AA07138; Wed, 5 Jul 89 12:10:14 PDT
Received: from east.berkeley.edu by RELAY.CS.NET id aa14887; 5 Jul 89 15:04 EDT
Received: by east.Berkeley.EDU (5.58/1.34)
    id AA28003; Wed, 5 Jul 89 11:03:26 PST
Message-Id: <[e-mail]>
Date: Wed, 5 Jul 89 11:03:24 PST
From: Phil Lapsley <[e-mail]>
To: [e-mail]
Subject: Ideas for non-modem DSP

I realize that there is a lot of interest in using DSP for amateur modem
applications, which is well and good.  But I'm also interested in
other applications of DSP towards ham radio.  For example:

1. Baseband spectrum analyzer; use is fairly obvious.

2. Digital audio storage scope, for analysis of strange sounds and
rtty/modem freqs, along with (1).

3. Adaptive audio filter.  For example, you tell your DSP filter that
you're having an SSB qso.  If somebody tunes up on top of you, it
automatically notches out the offending carrier.  Obviously, one
could also do a variable bandwidth audio CW filter.  (Such a shame
we can't do this at IF...)

4. (Back to modem type stuff).  Adaptive rtty/modem software.  Something
that could analyze the incoming signal and say, "Hurm, looks like baudot,
%d baud rate, a little bit off frequency..."

I realize that this requires a lot of filter design on the fly, but
I figure that's why you have the PC there: it can download stuff to
the DSP to analyze the signal, then the PC can compute a good filter,
and then download another program to the DSP to receive the signal
courtesy of the newly designed filter.

I figured I'd just toss this out as food for thought.  Do people have
other ideas?

                        Phil


From @relay.cs.net,@tektronix.tek.com:[e-mail]  Wed Jul  5 15:05:01 1989
Received: from flash.bellcore.com by ka9q.bellcore.com (5.60/4.7)
    id AA01243; Wed, 5 Jul 89 15:05:01 EST
Received: from RELAY.CS.NET by flash.bellcore.com (5.58/1.1)
    id AA17570; Wed, 5 Jul 89 16:05:42 EDT
Received: from tektronix.tek.com by RELAY.CS.NET id aa26008; 5 Jul 89 16:02 EDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA22942; Wed, 5 Jul 89 13:02:38 PDT
Received: by orca.wv.tek.com (5.51/7.1)
    id AA23877; Wed, 5 Jul 89 12:57:33 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA04283; Wed, 5 Jul 89 13:03:07 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA04281; Wed, 5 Jul 89 13:03:05 PDT
From: "Michael T. Horne" <[e-mail]>
Message-Id: <[e-mail]>
Date: Wed,  5 Jul 89 13:03:04 PDT
To: [e-mail]
Subject: Phil Lapsley's message

This is an instant replay of a message that Phil posted earlier today.  The
size of the list nearly tripled today, so about 2/3 of the people currently
on the list didn't see this message.  If you have already received this,
please ignore.

Mike

------- Forwarded Message
Date: Wed, 5 Jul 89 11:03:24 PST
>From: Phil Lapsley <[e-mail]>
To: [e-mail]
Subject: Ideas for non-modem DSP

I realize that there is a lot of interest in using DSP for amateur modem
applications, which is well and good.  But I'm also interested in
other applications of DSP towards ham radio.  For example:

1. Baseband spectrum analyzer; use is fairly obvious.

2. Digital audio storage scope, for analysis of strange sounds and
rtty/modem freqs, along with (1).

3. Adaptive audio filter.  For example, you tell your DSP filter that
you're having an SSB qso.  If somebody tunes up on top of you, it
automatically notches out the offending carrier.  Obviously, one
could also do a variable bandwidth audio CW filter.  (Such a shame
we can't do this at IF...)

4. (Back to modem type stuff).  Adaptive rtty/modem software.  Something
that could analyze the incoming signal and say, "Hurm, looks like baudot,
%d baud rate, a little bit off frequency..."

I realize that this requires a lot of filter design on the fly, but
I figure that's why you have the PC there: it can download stuff to
the DSP to analyze the signal, then the PC can compute a good filter,
and then download another program to the DSP to receive the signal
courtesy of the newly designed filter.

I figured I'd just toss this out as food for thought.  Do people have
other ideas?

                        Phil
------- End of Forwarded Message



From @relay.cs.net,@tektronix.tek.com:[e-mail]  Fri Jul  7 16:51:18 1989
Received: from flash.bellcore.com by ka9q.bellcore.com (5.60/4.7)
    id AA03659; Fri, 7 Jul 89 16:51:18 EST
Received: from relay.cs.net by flash.bellcore.com (5.58/1.1)
    id AA10035; Fri, 7 Jul 89 17:51:28 EDT
Received: from tektronix.tek.com by RELAY.CS.NET id aa02651; 7 Jul 89 17:47 EDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA23521; Fri, 7 Jul 89 14:40:30 PDT
Received: by orca.wv.tek.com (5.51/7.1)
    id AA01776; Fri, 7 Jul 89 14:36:02 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA08161; Fri, 7 Jul 89 14:44:07 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA08159; Fri, 7 Jul 89 14:44:04 PDT
From: "Michael T. Horne" <[e-mail]>
Message-Id: <[e-mail]>
Date: Fri,  7 Jul 89 14:44:03 PDT
To: [e-mail]
Subject: Nearing steady state...

The response from all of you has been tremendous!  I'm amazed at the amount
of interest in dsp (several of you mentioned projects that you are working
on; why not share with the rest of us?).

This message is really a bounce test, but just for fun, here's a list of
people so far that have subscribed to the dsp-group.  I expect that the
group will be pretty steady within the next week or so.

Mike
------------------------------------------------------------------------------
Currently subscribed to the dsp-group:

aa4cg![e-mail]        (Bernie ?)
[e-mail]                (Al Boehnlein)
[e-mail]                (Alan Biocca - WB6ZQZ)
[e-mail]                (Barry McLarnon - VE3JF)
[e-mail]                (Richard Bisbey - NG6Q)
[e-mail]            (Brian Shiratuki - N6SXT)
[e-mail]                (Bob Albrightson - N7AKR)
[e-mail]            (Maitland Bottoms - AA4HS)
[e-mail]                (Brian K. Moran - N9ADG)
[e-mail]                (Brian Lloyd - WB6RQN)
[e-mail]                (?)
[e-mail]            (Tom Clark - W3IWI)
[e-mail]                (Randy Cole - KN6W)
[e-mail]                (Doug Collinge)
[e-mail]                (Dave Borden - K8MMO)
[e-mail]            (Dave McNeill - WB7BLB)
[e-mail]                    (Don Merritt - K3BGE)
[e-mail]                    (?)
[e-mail]            (DSP archive pseudo-person)
[e-mail]            (Bdale's redist list)
[e-mail]            (David's redist list)
[e-mail]                (John's redist list)
[e-mail]    (Geert's redist list)
[e-mail]            (Phil's redist list)
[e-mail]        (Ray Gordon - G1XRN)
[e-mail]                    (Doug Faunt - N6TQS)
gatech!stiatl![e-mail]        (John DeArmond - WD4OQC)
[e-mail]          (Garry's redist list)
[e-mail]                (Greg Troxel - N1DAM)
[e-mail]            (Glen R. Sands - KA7AYF)
[e-mail]                (Harris Edelman - KB6OWB)
homxb![e-mail]                (Scott McLellan - ND3P)
[e-mail]                (Club Station)
[e-mail]            (Gary Huntress)
[e-mail]                (?)
integin![e-mail]            (Stephen Bechtolsheim - KX9G)
ism780c!kodak![e-mail]        (Barry L. Ornitz - WA4VZQ)
[e-mail]            (Jim English - WA7QLB)
[e-mail]                    (John B. Milton - N8KSN)
[e-mail]                (James Austin - KA2RVO)
[e-mail]                (Jeffrey Austen - K9JA)
[e-mail]                (John A. Limpert - N3DMC)
[e-mail]            (Kevin Burnett - N6UWS)
[e-mail]            (Jacque Kubley - KA9FJS)
[e-mail]            (Kaj Wiik - OH6EH)
[e-mail]            (Stephen Look - KA9SZW)
[e-mail]            (Bob McGwier - N4HY)
[e-mail]                (?)
[e-mail]                (Mark Matthews - WA6LZO)
[e-mail]                (Marc ? - KG6KF)
[e-mail]            (Michael Horne - KA7AXD)
microsoft![e-mail]    (Jeff Whitlatch - KO7M)
[e-mail]                (Michael Chace - G6DHU)
[e-mail]                (Jim Mankin - KB3KJ)
[e-mail]            (?)
[e-mail]                (Dave Trulli - NN2Z)
[e-mail]        (Ofer Lapid - 4X6OJ)
[e-mail]                (Dave Pascoe - KM3T)
[e-mail]            (Pat Davis - KD9UU)
[e-mail]                (Phil Lapsley - N6TCT)
[e-mail]                (Paul Shen)
[e-mail]                (Rusty Carruth - N7IKQ)
[e-mail]        (Jarkko Vuori)
sequent!islabs![e-mail]        (John Wipfli)
sequent![e-mail]            (Ken Dove - KF7RZ)
[e-mail]            (Stanley Dunn)
[e-mail]                (Steve H. Goldberg - WQ6L)
[e-mail]                (Larry Stewart - K4EO)
[e-mail]            (Mark Lomas)
ucsd.ucsd.edu!telesoft![e-mail]    (Bruce Bergman - N7HAW)
uunet!nikhefh.hep.nl![e-mail]    (Henk Peek - PA0HZP)
[e-mail]            (?)
[e-mail]                (?)
[e-mail]            (William Armitage)
cpoint![e-mail]            (Ken Wolff - K1EA)
[e-mail]            (?)



From @relay.cs.net,@tektronix.tek.com:[e-mail]  Sun Jul  9 18:55:58 1989
Received: from flash.bellcore.com by ka9q.bellcore.com (5.60/4.7)
    id AA06190; Sun, 9 Jul 89 18:55:58 EST
Received: from relay.cs.net by flash.bellcore.com (5.58/1.1)
    id AA07656; Sun, 9 Jul 89 19:56:38 EDT
Received: from tektronix.tek.com by RELAY.CS.NET id aa01243; 9 Jul 89 1:35 EDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA05760; Sat, 8 Jul 89 22:37:32 PDT
Received: by orca.wv.tek.com (5.51/7.1)
    id AA05994; Sat, 8 Jul 89 22:33:04 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA10351; Sat, 8 Jul 89 22:37:58 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA10349; Sat, 8 Jul 89 22:37:56 PDT
Received: by orca.wv.tek.com (5.51/7.1)
    id AA05924; Sat, 8 Jul 89 22:28:41 PDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA05703; Sat, 8 Jul 89 22:32:56 PDT
Received: from RELAY.CS.NET by zephyr.ENS.TEK.COM (4.0/7.1)
    id AA01347; Sat, 8 Jul 89 22:30:19 PDT
Received: from gte.com by RELAY.CS.NET id ab27622; 9 Jul 89 1:28 EDT
Received: from GODOT.DECnet MAIL11D_V3 by bunny.gte.com (4.41/GTEL1.11)
    id AA10602; Sun, 9 Jul 89 00:16:23 EDT
Date: Sun, 9 Jul 89 00:16:22 EDT
Message-Id: <[e-mail]>
From: DAVE PASCOE <[e-mail]>
To:
Mmdf-Warning:  Parse error in original version of preceding line at RELAY.CS.NET
Subject: Power line and atmospheric noise cancellation

I am wondering if anyone out there in the group has ever considered
using DSP for reducing power line noise and/or atmospheric noise.
I know that there are noise cancellation algorithms floating around
but I haven't had much time to take a look a them.

The benefit of reducing or eliminating power line noise in HF/VHF
radio receivers should be obvious.  Reducing atmospheric noise can
provide "low-band" DXers (160 and 80m) with improved S/N ratios thereby
improving their ability to hear weak signals.

I look forward to reading your responses.

***********************************************************************
*  Dave Pascoe          |     Internet: [e-mail]          *
*                       |         UUCP: ...!harvard!m2c!jjmhome!km3t  *
*  Day [phone]    | Smart Mailer: [e-mail]             *
*  Eve [phone]    | Packet Radio: KM3T @ K1UGM.MA.USA.NA        *
***********************************************************************




From @relay.cs.net,@tektronix.tek.com:[e-mail] Wed Jul  5 15:18:09 1989
Received: from flash.bellcore.com by ka9q.bellcore.com (5.60/4.7)
    id AA01196; Wed, 5 Jul 89 14:18:01 EST
Received: from RELAY.CS.NET by flash.bellcore.com (5.58/1.1)
    id AA14446; Wed, 5 Jul 89 15:18:41 EDT
Received: from tektronix.tek.com by RELAY.CS.NET id aa25166; 5 Jul 89 15:16 EDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA20804; Wed, 5 Jul 89 12:17:19 PDT
Received: by orca.wv.tek.com (5.51/7.1)
    id AA22750; Wed, 5 Jul 89 12:12:25 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA04160; Wed, 5 Jul 89 12:17:14 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA04158; Wed, 5 Jul 89 12:17:07 PDT
Received: by orca.wv.tek.com (5.51/7.1)
    id AA22691; Wed, 5 Jul 89 12:08:51 PDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA20627; Wed, 5 Jul 89 12:12:49 PDT
Received: from RELAY.CS.NET by zephyr.ENS.TEK.COM (4.0/7.1)
    id AA07138; Wed, 5 Jul 89 12:10:14 PDT
Received: from east.berkeley.edu by RELAY.CS.NET id aa14887; 5 Jul 89 15:04 EDT
Received: by east.Berkeley.EDU (5.58/1.34)
    id AA28003; Wed, 5 Jul 89 11:03:26 PST
Message-Id: <[e-mail]>
Date: Wed, 5 Jul 89 11:03:24 PST
From: Phil Lapsley <[e-mail]>
To: [e-mail]
Subject: Ideas for non-modem DSP
Status: RO

I realize that there is a lot of interest in using DSP for amateur modem
applications, which is well and good.  But I'm also interested in
other applications of DSP towards ham radio.  For example:

1. Baseband spectrum analyzer; use is fairly obvious.

2. Digital audio storage scope, for analysis of strange sounds and
rtty/modem freqs, along with (1).

3. Adaptive audio filter.  For example, you tell your DSP filter that
you're having an SSB qso.  If somebody tunes up on top of you, it
automatically notches out the offending carrier.  Obviously, one
could also do a variable bandwidth audio CW filter.  (Such a shame
we can't do this at IF...)

4. (Back to modem type stuff).  Adaptive rtty/modem software.  Something
that could analyze the incoming signal and say, "Hurm, looks like baudot,
%d baud rate, a little bit off frequency..."

I realize that this requires a lot of filter design on the fly, but
I figure that's why you have the PC there: it can download stuff to
the DSP to analyze the signal, then the PC can compute a good filter,
and then download another program to the DSP to receive the signal
courtesy of the newly designed filter.

I figured I'd just toss this out as food for thought.  Do people have
other ideas?

                        Phil


From @relay.cs.net,@tektronix.tek.com:[e-mail]  Mon Jul 10 15:19:11 1989
Received: from flash.bellcore.com by ka9q.bellcore.com (5.60/4.7)
    id AA07616; Mon, 10 Jul 89 15:19:11 EST
Received: from relay.cs.net by flash.bellcore.com (5.58/1.1)
    id AA16275; Mon, 10 Jul 89 16:19:57 EDT
Received: from tektronix.tek.com by RELAY.CS.NET id aa00568; 10 Jul 89 14:16 EDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA16635; Mon, 10 Jul 89 11:17:10 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA13113; Mon, 10 Jul 89 11:20:28 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA13111; Mon, 10 Jul 89 11:20:21 PDT
Received: by orca.wv.tek.com (5.51/7.1)
    id AA15959; Mon, 10 Jul 89 11:10:21 PDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA16511; Mon, 10 Jul 89 11:14:30 PDT
Received: from tektronix by zephyr.ENS.TEK.COM (4.0/7.1)
    id AB13630; Mon, 10 Jul 89 11:11:46 PDT
Received: from uunet.uu.net by RELAY.CS.NET id aa12260; 10 Jul 89 14:05 EDT
Received: from hp4nl.nluug.nl by uunet.uu.net (5.61/1.14) with SMTP
    id AA06030; Mon, 10 Jul 89 10:19:53 -0400
Received: from [192.16.183.3] by hp4nl.nluug.nl with SMTP
          id AA11445 (5.58.1.14/2.14); Mon, 10 Jul 89 16:01:52 MET
Received: by nikhefh.hep.nl; Mon, 10 Jul 89 16:02:20 +0200 (MET)
Message-Id: <[e-mail]>
Date: Mon, 10 Jul 89 16:02:20 +0200
From: Henk Peek <[e-mail]>
To: [e-mail], [e-mail]
Subject: Re:  Ideas for non-modem DSP


>I realize that there is a lot of interest in using DSP for amateur modem
>applications, which is well and good.  But I'm also interested in
>other applications of DSP towards ham radio.  For example:
>1. Baseband spectrum analyzer; use is fairly obvious.
>2. Digital audio storage scope, for analysis of strange sounds and
>3. Adaptive audio filter.  For example, you tell your DSP filter that
>4. (Back to modem type stuff).  Adaptive rtty/modem software.
>I figured I'd just toss this out as food for thought.  Do people have
>other ideas?
Yes, many.... Some are:

Measurement:
  Audio signal generator
  Function generator
  Arbitrary function generator
  Stereo signal generator and demodulator
  Wow and flutter meter
  Phase meter

Transmitter:
  Intelligent modulation compressor
  Automatic nonlinear modulation signal expander to compensate PA compression
  Generatie of 90 degree shifted audio signals (phase SSB)

Receiver:
  Intelligent AVC generation
  Variable MF bandwidth filter (It isn't possible to do all the MF filtering by
                a DSP. (dynamic range of the ADC :-( )
  Intelligent impulse noise gate
  Many types of demodulator (SSB, FM, packet etc.)
  Audio filters


When you combine many of those, you can make a transceiver with more bells
than the latest highend ICOM models :-). The trick is that you put so many
functions as possible in a DSP. Then can *you* make the  radio with not so
many components as the current one's.  The next generation transceiver??


Henk, PA0HZP   ..!uunet!mcvax!nikhefk!henkp.UUCP



From @relay.cs.net,@tektronix.tek.com:[e-mail]  Mon Jul 10 17:02:05 1989
Received: from flash.bellcore.com by ka9q.bellcore.com (5.60/4.7)
    id AA07691; Mon, 10 Jul 89 17:02:05 EST
Received: from relay.cs.net by flash.bellcore.com (5.58/1.1)
    id AA23403; Mon, 10 Jul 89 18:02:41 EDT
Received: from tektronix.tek.com by RELAY.CS.NET id aa03226; 10 Jul 89 16:39 EDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA22182; Mon, 10 Jul 89 13:38:53 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA13520; Mon, 10 Jul 89 13:40:08 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA13515; Mon, 10 Jul 89 13:39:53 PDT
From: "Michael T. Horne" <[e-mail]>
Message-Id: <[e-mail]>
Date: Mon, 10 Jul 89 13:39:52 PDT
To: Andrew Elms <[e-mail]>
Cc: [e-mail]
Subject: Re: text on MSK/QPSK design
In-Reply-To: Your message of Mon, 10 Jul 89 15:18:55 BST.
             <[e-mail]>


>Also, can you refer me to any suitable texts for a project
>on an IC implementation of an MSK or QPSK decoder.
>
>Thanks ( in advance ),
>Andy Elms ( [e-mail] )

There are many books on the fundamentals of data communications, and you might
be able to preen some information from the many flyers/propaganda that the
modem chip manufacturers print.  I know of several application notes that
TI, Motorola, and others have printed on the implementation of modems on their
DSP chips.  TI has a very good app note describing the implementation of a
Bell 212 modem on their TMS320 series parts; it gives a pretty thorough
description of the workings of Bell 212 and DPSK modems in general.  I don't
have the app note number handy, but if needed I can find it and send it to
you.

Why not use a DSP chip instead of lots of discrete digital parts?  What are
your speed requirements?

Can anyone else recommend a good reference?

Mike



From @relay.cs.net,@tektronix.tek.com:[e-mail]  Mon Jul 10 20:15:29 1989
Received: from flash.bellcore.com by ka9q.bellcore.com (5.60/4.7)
    id AA07805; Mon, 10 Jul 89 20:15:29 EST
Received: from relay.cs.net by flash.bellcore.com (5.58/1.1)
    id AA27837; Mon, 10 Jul 89 21:16:01 EDT
Received: from tektronix.tek.com by RELAY.CS.NET id aa06715; 10 Jul 89 20:52 EDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA01722; Mon, 10 Jul 89 17:53:04 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA14060; Mon, 10 Jul 89 17:55:43 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA14056; Mon, 10 Jul 89 17:55:38 PDT
Received: by orca.wv.tek.com (5.51/7.1)
    id AA26832; Mon, 10 Jul 89 17:45:46 PDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA01626; Mon, 10 Jul 89 17:50:02 PDT
Received: from RELAY.CS.NET by zephyr.ENS.TEK.COM (4.0/7.1)
    id AA18033; Mon, 10 Jul 89 17:47:20 PDT
Received: from east.berkeley.edu by RELAY.CS.NET id aa18519; 10 Jul 89 20:42 EDT
Received: by east.Berkeley.EDU (5.58/1.34)
    id AA05000; Mon, 10 Jul 89 15:26:16 PST
Message-Id: <[e-mail]>
Date: Mon, 10 Jul 89 15:26:14 PST
From: Phil Lapsley <[e-mail]>
To: "Michael T. Horne" <[e-mail]>,
        Andrew Elms <[e-mail]>
Subject: Re: text on MSK/QPSK design
Cc: [e-mail]

Silicon Systems just co-produced a book on modem design by one of the
big names in early modem efforts (forget who, sorry).  Send me mail if
you want their address if you can't get it on your own.  (I'm on
vacation in San Diego right now, so I'm not around my notes.)

                    Phil


From @relay.cs.net,@tektronix.tek.com:[e-mail]  Tue Jul 11 06:51:52 1989
Received: from flash.bellcore.com by ka9q.bellcore.com (5.60/4.7)
    id AA08561; Tue, 11 Jul 89 06:51:52 EST
Received: from relay.cs.net by flash.bellcore.com (5.58/1.1)
    id AA12007; Tue, 11 Jul 89 07:52:37 EDT
Received: from tektronix.tek.com by RELAY.CS.NET id aa25259; 11 Jul 89 7:52 EDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA15459; Tue, 11 Jul 89 04:53:44 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA14752; Tue, 11 Jul 89 04:58:12 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA14750; Tue, 11 Jul 89 04:58:09 PDT
Received: by orca.wv.tek.com (5.51/7.1)
    id AA12008; Tue, 11 Jul 89 04:48:06 PDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA15413; Tue, 11 Jul 89 04:52:20 PDT
Received: from RELAY.CS.NET by zephyr.ENS.TEK.COM (4.0/7.1)
    id AA01847; Tue, 11 Jul 89 04:49:37 PDT
Received: from uunet.uu.net by RELAY.CS.NET id aa24170; 11 Jul 89 7:47 EDT
Received: from ukc.UUCP by uunet.uu.net (5.61/1.14) with UUCP
    id AA10055; Tue, 11 Jul 89 07:46:44 -0400
Received: from nw.stl.stc.co.uk by kestrel.Ukc.AC.UK   via PSS (UKC CAMEL FTP)
           id aa06712; 11 Jul 89 12:15 BST
Via: nw.stl.stc.co.uk; Tue, 11 Jul 89 12:13:39 BST
Received: from hilbert.praxis.co.uk (hilbert) by newton.praxis.co.uk; Tue, 11 Jul 89 09:36:31 GMT
From: Andrew Elms <tektronix.TEK.COM!orca.WV.TEK.COM!tektronix.TEK.COM!uunet.uu.net!zephyr.ENS.TEK.COM!praxis![e-mail]>
Message-Id: <[e-mail]>
To: [e-mail]
Date: Tue, 11 Jul 89 9:40:54 BST
Subject: Re: text on MSK/QPSK design
X-Mailer: Elm [version 1.5c]

In reply to Mike Horne's help about MSK/QPSK decoders.


>
> > Why not use a DSP chip instead of lots of discrete digital parts?  What are
> > your speed requirements?
> >
> > Mike
> >
>
> Mike,
>
> Thanks for your help. The answer to the question is that I am to be
> doing a _custom_ design of an MSK or QPSK demodulator. This is for a
> project at the University of Surrey (starting in October) as part of my
> Master of Engineering degree in E & EE.
>
> The design will be in Ella ( a hardware design and description language )
> and will be synthesised from high level Ella by Locam ( a logic
> synthesis tool ) onto an ASIC cell library.
>
> My query is not so much one of the communications theory, more
> of implementation details - hence any references in this direction
> are most appreciated.
>
> Any thoughts ?
>
> Andy
>
> --------------------------------------------------------------------------
> | Andrew Elms.                          | E-Mail: ([e-mail]) |
> | Praxis Electronic Design              | Tel   : [phone]         |
> | 20 Manvers Street, Bath.              | Fax   : [phone]         |
> | BA1 1PX, U.K.                         | Telex : 445848 PRAXIS G        |
> --------------------------------------------------------------------------
>
>
>

Can anyone else offer any assistance?

Thanks

Andy.


From @relay.cs.net,@tektronix.tek.com:[e-mail]  Tue Jul 11 14:03:05 1989
Received: from flash.bellcore.com by ka9q.bellcore.com (5.60/4.7)
    id AA08929; Tue, 11 Jul 89 14:03:05 EST
Received: from relay.cs.net by flash.bellcore.com (5.58/1.1)
    id AA15623; Tue, 11 Jul 89 15:03:47 EDT
Received: from tektronix.tek.com by RELAY.CS.NET id aa09694; 11 Jul 89 15:03 EDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA29930; Tue, 11 Jul 89 12:03:53 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA15599; Tue, 11 Jul 89 12:07:08 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA15597; Tue, 11 Jul 89 12:07:04 PDT
Received: by orca.wv.tek.com (5.51/7.1)
    id AA25242; Tue, 11 Jul 89 11:56:51 PDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA29701; Tue, 11 Jul 89 12:01:03 PDT
Received: from tektronix by zephyr.ENS.TEK.COM (4.0/7.1)
    id AB05551; Tue, 11 Jul 89 11:58:22 PDT
Received: from cgl.ucsf.edu by RELAY.CS.NET id aa28817; 11 Jul 89 14:53 EDT
Received: by cgl.ucsf.EDU (5.59/GSC4.16)
    id AA15400; Tue, 11 Jul 89 11:51:09 PDT
Received: from noe.UUCP by hop.toad.com id AA07059; Tue, 11 Jul 89 11:02:15 PDT
Received: by noe.UUCP (KG6KF's qsmail2.3)
    id AA12566; 11 Jul 89 09:46:34 PDT (Tue)
To: [e-mail]
Date: Tue, 11 Jul 89 9:46:27 PDT
Subject: Re: more ideas
X-Mailer: Elm [version 1.5]
From: [e-mail]
Mmdf-Warning:  Parse error in original version of preceding line at RELAY.CS.NET
Message-Id: <[e-mail]>

Some ideas I've had for non-modem DSP applications:

- Intelligent compansion/filtering of SSB with a low-bandwidth digital control
  channel superimposed so the DSP chips can talk.

- Simple auto-correlation filters to filter atmospheric noise for
  CW operation.

- "Predictive filtering" -- I haven't played with this in years --
  the use of "acausal" filters or filters whose output depends on
  the FUTURE input as well as the past input.  Coupled with a fast
  CPU to churn out impulse responses, it could be used to remove
  slowly varying noise signals (such as spark plug popping which
  changes rate as the engine speed changes).  It can be used to
  increase speed of response to a change in the input signal,
  much like the hardware technique of "feed-forward".

- Dynamic zero-redundancy digital audio compression.

- Frequency-compandored SSB.

- Coherent CW -- a technique whereby digital filtering is used to
  reduce the received bandwidth to around 1 Hz.  Because the
  received band is so narrow, there is practically no noise received.
  Has been used to carry on HF QSO's between California and Japan with
  100 mW of power.  Reliably.

Marc  KG6KF
[e-mail]


From @relay.cs.net,@tektronix.tek.com:[e-mail]  Tue Jul 11 18:16:56 1989
Received: from flash.bellcore.com by ka9q.bellcore.com (5.60/4.7)
    id AA09083; Tue, 11 Jul 89 18:16:56 EST
Received: from relay.cs.net by flash.bellcore.com (5.58/1.1)
    id AA03391; Tue, 11 Jul 89 19:17:42 EDT
Received: from tektronix.tek.com by RELAY.CS.NET id aa12253; 11 Jul 89 16:49 EDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA03442; Tue, 11 Jul 89 13:49:31 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA15809; Tue, 11 Jul 89 13:50:07 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA15802; Tue, 11 Jul 89 13:49:56 PDT
From: "Michael T. Horne" <[e-mail]>
Message-Id: <[e-mail]>
Date: Tue, 11 Jul 89 13:49:54 PDT
To: [e-mail]
Cc: [e-mail]
Subject: Re: more ideas
In-Reply-To: Your message of Tue, 11 Jul 89 9:46:27 PDT.
             <[e-mail]>



>   Coherent CW -- a technique whereby digital filtering is used to
>   reduce the received bandwidth to around 1 Hz.  Because the
>   received band is so narrow, there is practically no noise received.
>   Has been used to carry on HF QSO's between California and Japan with
>   100 mW of power.  Reliably.

I have some questions WRT this.  The KTB noise floor for a 1 Hz filter is
easy to accept (~ -174 dBm, 50 ohm system, etc.).  However, for a filter of
this bandwidth, the charge time is very long, i.e. your keying rate has to
be *very* slow.  You can have a wonderful signal-to-noise ratio, but due to
the bandwidth limitation, your information transfer rate will be very
low.  Shannon strikes again! :)

Now I'm not familiar with coherent CW, but if it's like any other coherent
detection problem, clocking (or other) information must be present on the
decoding end.  However, I don't see how this will buy you anything if you're
working in a 1 Hz bandwidth (perhaps you have a better BER, but with a
noise floor that low, it's probably good enough as it is).

Could you elaborate more, Marc (or anyone else)?

Mike



From @relay.cs.net,@tektronix.tek.com:[e-mail]  Tue Jul 11 20:18:20 1989
Received: from flash.bellcore.com by ka9q.bellcore.com (5.60/4.7)
    id AA09203; Tue, 11 Jul 89 20:18:20 EST
Received: from relay.cs.net by flash.bellcore.com (5.58/1.1)
    id AA08788; Tue, 11 Jul 89 21:19:03 EDT
Received: from tektronix.tek.com by RELAY.CS.NET id aa16269; 11 Jul 89 21:14 EDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA12897; Tue, 11 Jul 89 18:11:22 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA16407; Tue, 11 Jul 89 18:11:48 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA16400; Tue, 11 Jul 89 18:11:37 PDT
From: "Michael T. Horne" <[e-mail]>
Message-Id: <[e-mail]>
Date: Tue, 11 Jul 89 18:11:36 PDT
To: [e-mail]
Cc: [e-mail]
Subject: Is it conference time yet?
Fcc: outbox


Ok Bdale, et. al., what's the scoop on the ARRL conference ?  I think we
all know when it is, but what about accomodations, pre-functions, etc.? :)
If something was already posted regarding these questions, could someone
rehash them for me?

How many of you folks are planning on writing or presenting tcp/ip or
dsp-related papers this year?  Do you think tcp/ip/dsp will be able to
`take over' the conference this year? ;-)

Mike



From @relay.cs.net,@tektronix.tek.com:[e-mail]  Wed Jul 12 06:48:07 1989
Received: from flash.bellcore.com by ka9q.bellcore.com (5.60/4.7)
    id AA09722; Wed, 12 Jul 89 06:48:07 EST
Received: from relay.cs.net by flash.bellcore.com (5.58/1.1)
    id AA23548; Wed, 12 Jul 89 07:48:57 EDT
Received: from tektronix.tek.com by RELAY.CS.NET id aa10898; 12 Jul 89 7:46 EDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA26699; Wed, 12 Jul 89 04:45:45 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA17145; Wed, 12 Jul 89 04:47:29 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA17143; Wed, 12 Jul 89 04:47:27 PDT
Received: by orca.wv.tek.com (5.51/7.1)
    id AA20124; Wed, 12 Jul 89 04:37:06 PDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA26650; Wed, 12 Jul 89 04:41:23 PDT
Received: from tektronix by zephyr.ENS.TEK.COM (4.0/7.1)
    id AB13515; Wed, 12 Jul 89 04:38:39 PDT
Received: from cgl.ucsf.edu by RELAY.CS.NET id aa09182; 12 Jul 89 7:38 EDT
Received: by cgl.ucsf.EDU (5.59/GSC4.16)
    id AA02996; Wed, 12 Jul 89 04:37:25 PDT
Received: from noe.UUCP by hop.toad.com id AA16429; Wed, 12 Jul 89 04:37:05 PDT
Received: by noe.UUCP (KG6KF's qsmail2.3)
    id AA15525; 12 Jul 89 04:24:07 PDT (Wed)
Received: by noe.UUCP (KG6KF's qsmail2.3)
    id AA15506; 12 Jul 89 04:21:14 PDT (Wed)
To: "Michael T. Horne" <[e-mail]>
Date: Wed, 12 Jul 89 4:21:08 PDT
Subject: Re: coherent CW
In-Reply-To: Message from "Michael T. Horne" of Jul 11, 89 at 1:49 pm
X-Mailer: Elm [version 1.5]
From: [e-mail]
Mmdf-Warning:  Parse error in original version of preceding line at RELAY.CS.NET
Message-Id: <[e-mail]>

Mike,
You are right -- coherent CW is slow.  Of course, the DSP chip can
dynamically adjust the rate... :-)

The work on coherent CW was done by Charles "Woody" Woodson who is
a ham (sorry, forgot the call) and is a professor at UC Berkeley.
He may be on the net -- could someone try to finger him?

His last name may be Woodsen.

-Marc KG6KF



From @relay.cs.net,@tektronix.tek.com:[e-mail]  Wed Jul 12 17:20:05 1989
Received: from flash.bellcore.com by ka9q.bellcore.com (5.60/4.7)
    id AA10316; Wed, 12 Jul 89 17:20:05 EST
Received: from relay.cs.net by flash.bellcore.com (5.58/1.1)
    id AA10056; Wed, 12 Jul 89 18:20:57 EDT
Received: from tektronix.tek.com by RELAY.CS.NET id aa02260; 12 Jul 89 18:20 EDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA18276; Wed, 12 Jul 89 15:21:24 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA18268; Wed, 12 Jul 89 15:22:23 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA18265; Wed, 12 Jul 89 15:22:21 PDT
From: "Michael T. Horne" <[e-mail]>
Message-Id: <[e-mail]>
Date: Wed, 12 Jul 89 15:22:19 PDT
To: [e-mail]
Subject: Re: coherent CW and 1Hz filters
In-Reply-To: Your message of Wed, 12 Jul 89 4:21:08 PDT.
             <[e-mail]>


And another thing... :)

For a nice, narrow (1Hz) filter, you're going to need a fairly large number
of filter coefficients (if you implement an FIR filter), or many poles of
a simpler IIR filter, perhaps with wide working registers.  For an FIR
filter, this translates to heavy computation time, as well as lengthy phase
delay.

Also, to be able to distinguish signals deep down `in the noise', you'll need
1) a fairly small A/D reference voltage, 2) a wide A/D (if you want any sort
of dynamic range), 3) a DAMN good power supply (heavy filtering, low
noise, etc.), and 4) excellent layout/design techniques (with low S/N ratios
in a system like this, we're talking *very* small signals).

>The work on coherent CW was done by Charles "Woody" Woodson who is
>a ham (sorry, forgot the call) and is a professor at UC Berkeley.
>He may be on the net -- could someone try to finger him?

Anybody know this guy?  Perhaps he could be coaxed into joining the group!

Mike



From @relay.cs.net,@tektronix.tek.com:[e-mail]  Wed Jul 12 19:53:02 1989
Received: from flash.bellcore.com by ka9q.bellcore.com (5.60/4.7)
    id AA10519; Wed, 12 Jul 89 19:53:02 EST
Received: from relay.cs.net by flash.bellcore.com (5.58/1.1)
    id AA13844; Wed, 12 Jul 89 20:53:50 EDT
Received: from tektronix.tek.com by RELAY.CS.NET id aa06649; 12 Jul 89 20:51 EDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA22433; Wed, 12 Jul 89 17:52:29 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA18624; Wed, 12 Jul 89 17:57:03 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA18622; Wed, 12 Jul 89 17:57:00 PDT
Received: by orca.wv.tek.com (5.51/7.1)
    id AA17113; Wed, 12 Jul 89 17:46:22 PDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA22361; Wed, 12 Jul 89 17:50:35 PDT
Received: from ka9q.bellcore.com by RELAY.CS.NET id aa18603; 12 Jul 89 20:39 EDT
Received: by ka9q.bellcore.com (5.60/4.7)
    id AA10481; Wed, 12 Jul 89 19:37:02 EST
Date: Wed, 12 Jul 89 19:37:02 EST
From: Phil Karn <[e-mail]>
Message-Id: <[e-mail]>
To: [e-mail]
Subject: Re: coherent CW and 1Hz filters

A search of my online callbook for "charles woodson" turned up W6NEY,
who indeed lives in Berkeley, CA.

I think the "coherent CW" concept is useful mainly as a tool for demonstrating
the power of DSP to hams. There are, however, much more effective modulation
and coding techniques available if you actually want to establish a
functioning communication system with very weak signals.

Phil


From @relay.cs.net,@tektronix.tek.com:[e-mail]  Wed Jul 12 20:27:52 1989
Received: from flash.bellcore.com by ka9q.bellcore.com (5.60/4.7)
    id AA10571; Wed, 12 Jul 89 20:27:52 EST
Received: from relay.cs.net by flash.bellcore.com (5.58/1.1)
    id AA14458; Wed, 12 Jul 89 21:28:44 EDT
Received: from tektronix.tek.com by RELAY.CS.NET id aa07551; 12 Jul 89 21:26 EDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA23591; Wed, 12 Jul 89 18:27:25 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA18788; Wed, 12 Jul 89 18:31:57 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA18786; Wed, 12 Jul 89 18:31:55 PDT
From: "Michael T. Horne" <[e-mail]>
Message-Id: <[e-mail]>
Date: Wed, 12 Jul 89 18:31:54 PDT
To: [e-mail]
Subject: C implementation of the Remez-Exchange algorithm?
Fcc: outbox

I have the FORTAN implementation of the filter design program written
by Parks, McClellan, and Rabiner, but I would like to obtain a C version
of it, preferably in the public domain.  Anyone have such a beast?

Mike



From @relay.cs.net,@tektronix.tek.com:[e-mail]  Thu Jul 13 20:30:38 1989
Received: from flash.bellcore.com by ka9q.bellcore.com (5.60/4.7)
    id AA11630; Thu, 13 Jul 89 20:30:38 EST
Received: from relay.cs.net by flash.bellcore.com (5.58/1.1)
    id AA01134; Thu, 13 Jul 89 21:31:13 EDT
Received: from tektronix.tek.com by RELAY.CS.NET id aa08155; 13 Jul 89 16:47 EDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA24702; Thu, 13 Jul 89 13:49:00 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA20545; Thu, 13 Jul 89 13:54:15 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA20543; Thu, 13 Jul 89 13:54:12 PDT
Received: by orca.wv.tek.com (5.51/7.1)
    id AA16447; Thu, 13 Jul 89 13:43:17 PDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA24640; Thu, 13 Jul 89 13:47:35 PDT
Received: from tektronix by zephyr.ENS.TEK.COM (4.0/7.1)
    id AB05865; Thu, 13 Jul 89 13:44:52 PDT
Received: from east.berkeley.edu by RELAY.CS.NET id aa03653; 13 Jul 89 16:42 EDT
Received: by east.Berkeley.EDU (5.58/1.34)
    id AA25624; Thu, 13 Jul 89 09:11:10 PST
Message-Id: <[e-mail]>
Date: Thu, 13 Jul 89 09:11:08 PST
From: Phil Lapsley <[e-mail]>
To: [e-mail]
Subject: Book on modem design

I've received several queries about the book on modem design that I
mentioned.  I'm back from vacation now (sigh), so I just called
Silicon Systems.  The book is:

    Theory and Practice of Modem Design
    John A. C. Bingham
    Published by Wiley Interscience
    Rumored telephone number of Wiley: [phone]

Caveat: I haven't read the book, so I can't say how good it is.
I recall from the postcard they sent me that it did have some
stuff on QPSK designs.

Silicon Systems is sponsoring a traveling two day seminar by Bingham
on modem design.  If you want more information on this, call Fred
Kamp at SSI: [phone], x3202.

FYI: Silicon Systems is a producer of communications ICs.  Their
1988 product line consisted of single chip modems, modem front
ends, modem filters, DTMF receivers and transceivers, crosspoint
switches, and PCM/T1 interfaces.  You can get a databook on
their products from the above (714) number.

Of course, I have no relation to these guys.  I'm not even a customer!

                        Phil


From @relay.cs.net,@tektronix.tek.com:[e-mail]  Thu Jul 13 20:39:53 1989
Received: from flash.bellcore.com by ka9q.bellcore.com (5.60/4.7)
    id AA11642; Thu, 13 Jul 89 20:39:53 EST
Received: from relay.cs.net by flash.bellcore.com (5.58/1.1)
    id AA01532; Thu, 13 Jul 89 21:40:39 EDT
Received: from tektronix.tek.com by RELAY.CS.NET id aa11755; 13 Jul 89 20:22 EDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA01087; Thu, 13 Jul 89 17:23:48 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA20880; Thu, 13 Jul 89 17:26:22 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA20877; Thu, 13 Jul 89 17:26:18 PDT
Received: by orca.wv.tek.com (5.51/7.1)
    id AA24975; Thu, 13 Jul 89 17:15:22 PDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA01016; Thu, 13 Jul 89 17:19:44 PDT
Received: from tektronix by zephyr.ENS.TEK.COM (4.1/7.1)
    id AB00661; Thu, 13 Jul 89 17:16:58 PDT
Received: from east.berkeley.edu by RELAY.CS.NET id aa07823; 13 Jul 89 20:12 EDT
Received: by east.Berkeley.EDU (5.58/1.34)
    id AA29545; Thu, 13 Jul 89 13:12:28 PST
Message-Id: <[e-mail]>
Date: Thu, 13 Jul 89 13:12:25 PST
From: Phil Lapsley <[e-mail]>
To: "Michael T. Horne" <[e-mail]>
Subject: Re: coherent CW and 1Hz filters
Cc: [e-mail]

> >The work on coherent CW was done by Charles "Woody" Woodson who is
> >a ham (sorry, forgot the call) and is a professor at UC Berkeley.
> >He may be on the net -- could someone try to finger him?
>
> Anybody know this guy?  Perhaps he could be coaxed into joining the group!

Yeah, he's the faculty trustee for the U.C. Amateur Radio Club, W6BB,
of which I was President last year.  He's on the net in the sense that
he has an email address, but he doesn't read his mail, and very rarely
logs in.  He's pretty busy with a bunch of things related to a business
he recently inherited, and doesn't have much time to do ham radio these
days.

                        Phil


From @relay.cs.net,@tektronix.tek.com:[e-mail] Thu Jun 29 15:30:41 1989
Received: from flash.bellcore.com by ka9q.bellcore.com (5.60/4.7)
    id AA14529; Thu, 29 Jun 89 14:30:29 EST
Received: from RELAY.CS.NET by flash.bellcore.com (5.58/1.1)
    id AA25694; Thu, 29 Jun 89 15:30:57 EDT
Received: from tektronix.tek.com by RELAY.CS.NET id aa24649; 29 Jun 89 14:52 EDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA22595; Thu, 29 Jun 89 11:53:36 PDT
Received: by orca.wv.tek.com (5.51/7.1)
    id AA23790; Thu, 29 Jun 89 11:49:29 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA21685; Thu, 29 Jun 89 11:55:28 PDT
From: "Michael T. Horne" <[e-mail]>
Message-Id: <[e-mail]>
Date: Thu, 29 Jun 89 11:55:26 PDT
To: Phil Karn <[e-mail]>
Cc: [e-mail], [e-mail]
Subject: Re:  Designing with MF-n's
In-Reply-To: Your message of Thu, 29 Jun 89 01:32:31 EST.
             <[e-mail]>
Status: RO


(Note the Cc:  Let's move similar discussions to the new dsp-group!:)

>As you point out, aliasing can be a problem. (BTW, the switched capacitor
>filter is still an analog device, even though it works with discrete time
>samples; it does not quantize its samples).  Aliasing is not normally a
>problem with a fixed clock frequency; the 50x or 100x clock makes it
>possible to use simple RC filters.  But you may run into troubles if you
>want to steer the clock over a wide range because you may need to track the
>external anti-alias and reconstruction filters as well.

True.  The SCF belongs in the "discrete-time, continuous amplitude" category
of systems.  Nevertheless, the stepped output introduces problems that
may be difficult to rectify.

My intentions are to use the parts as programmable anti-aliasing and smoothing
filters on the input/output of an ADC/DAC pair.  Aliasing due to frequencies
at 1/2fclk and above may cause problems for the ADC input; however, using the
100X clock puts the Nyquist rate at a frequency I can pretty much guarantee
won't show up on the input.  As for the DAC, aliasing problems occur due to
high frequency energy at the sample transitions (due to steps in the output).
There is also a problem with glitches appearing in the output of the DAC when
you toggle a significantly large number of input pins (e.g. when you go from
0x7f to 0x80 on an 8-bit DAC, you are changing 8 `switches', and your
output could go to zero or Vref for a very short period of time.  Some DACs
attempt to correct for this, but not all do).  My concern is in significantly
strong frequency components due to these problems occuring at frequencies
greater than 1/2fclk, and the more bits in your DAC, the wider your dynamic
range for transmitting this components to the filter.

As far as output ripple due to the time quantization (more correct, Phil? :),
just filtering it with an RC won't work for my application, since the RC
filter will introduce non-linear phase distortions, one of the things I'm
trying to eliminate.  Using the 100X clock may reduce this problem
significantly, but it won't eliminate it.

Where's my programmable, true-analog filter?!? :)

>Another problem with these switched capacitor filters is dynamic range
>(i.e., noise) and DC offset. The latter, which varies with clock frequency,
>can be a problem with PSK demodulators if you're not careful in the setting
>of the slicer threshold.

Yes, the 100x part introduces nearly .5V in DC offset.  Another problem that
these parts have is their harmonic distortion.  If you can even find a spec,
it's usually around -57 to -60 dBc, and some quick math shows you that with a
12-bit ADC, these distortion components can easily be detected by the ADC.

>But in this day and age, why build anything but very high speed modems from
>discrete components? Use a DSP system! :-)

BINGO!  Give that man a cigar!  All of these `modern' modems can easily be
implemented on a 56K based board (nevermind the newer parts e.g. 96K, C30, 32C,
etc.).  I've even begun to think that I could implement the 56kbps MSK modem
on the 56K (short, of course, of the IF generation :)!  But if you want to
build a system that uses many different sampling rates (i.e. versatile), it
would be nice to have programmable anti-alising/smoothing filters on the
input/output (hence my rabid discussion about the MF series parts).

Mike



From @relay.cs.net,@tektronix.tek.com:[e-mail] Thu Jun 29 18:43:38 1989
Received: from flash.bellcore.com by ka9q.bellcore.com (5.60/4.7)
    id AA14696; Thu, 29 Jun 89 17:43:30 EST
Received: from RELAY.CS.NET by flash.bellcore.com (5.58/1.1)
    id AA13731; Thu, 29 Jun 89 18:44:00 EDT
Received: from tektronix.tek.com by RELAY.CS.NET id aa28151; 29 Jun 89 17:00 EDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA25979; Thu, 29 Jun 89 13:42:08 PDT
Received: by orca.wv.tek.com (5.51/7.1)
    id AA26942; Thu, 29 Jun 89 13:38:01 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA21970; Thu, 29 Jun 89 13:43:30 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA21968; Thu, 29 Jun 89 13:43:22 PDT
From: "Michael T. Horne" <[e-mail]>
Message-Id: <[e-mail]>
Date: Thu, 29 Jun 89 13:43:20 PDT
To: Phil Lapsley <[e-mail]>
Cc: [e-mail]
Subject: Re: DSP 56k
In-Reply-To: Your message of Wed, 28 Jun 89 13:06:42 PST.
             <[e-mail]>
Status: RO

>Earlier this summer I bought an Ariel PC-56 board for my PC.
>This has a Motorola DSP56001, a 14 bit AD/DA (TI chip, the TLC32014,
>I think), 24 lines of parallel I/O, and 24k words (24 bit wide) of
>fast RAM.

I've been considering buying one of these for a task at work, but currently
have my hands full for a while (though I may still pick one up).  I am curious
about what the Ariel board uses for anti-aliasing and smoothing filters on
the input/output to the AD/DA.  Do you have a schematic, Phil?

One thing that bothers me with a 14-bit ADC on that board is that it is
obviously using the PC supplies (which I suspect are far from clean).
I wonder how much feedthrough there is, and whether the ripple on the
supply lines get into the acquisition hardware.  Remember that a 14-bit
ADC can quantize a 10V range down to .61 mV!

>Also, does anybody have any neat 56k software they'd like to part with?

There exists a whizzy program that turns your favorite scope into a baseband
spectrum analyzer.  It samples data at a sufficiently high rate, transforms
it into the Fourier domain, takes the log of the data, then writes it out
a DAC at a rate calibrated with the sweep rate of your scope.  I haven't
seen it in operation, but it sounds like a cute program. :)

Mike



From @relay.cs.net,@tektronix.tek.com:[e-mail]  Wed Jul 26 23:26:04 1989
Received: from [128.96.32.20] by ka9q.bellcore.com (5.60/4.7)
    id AA15268; Wed, 26 Jul 89 23:26:04 EST
Received: from relay.cs.net by flash.bellcore.com (5.58/1.1)
    id AA26459; Wed, 26 Jul 89 12:59:54 EDT
Received: from tektronix.tek.com by RELAY.CS.NET id aa27098; 26 Jul 89 12:56 EDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA22877; Wed, 26 Jul 89 09:56:01 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA04119; Wed, 26 Jul 89 10:02:01 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA04116; Wed, 26 Jul 89 10:01:57 PDT
Received: by orca.wv.tek.com (5.51/7.1)
    id AA17299; Wed, 26 Jul 89 09:49:16 PDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA22780; Wed, 26 Jul 89 09:53:33 PDT
Received: from tektronix.tek.com by zephyr.ENS.TEK.COM (4.1/7.1)
    id AB21019; Wed, 26 Jul 89 09:51:04 PDT
Received: from east.berkeley.edu by RELAY.CS.NET id aa09789; 26 Jul 89 12:48 EDT
Received: by east.Berkeley.EDU (5.58/1.34)
    id AA21532; Wed, 26 Jul 89 08:47:57 PST
Message-Id: <[e-mail]>
Date: Wed, 26 Jul 89 08:47:56 PST
From: Phil Lapsley <[e-mail]>
To: "Michael T. Horne" <[e-mail]>,
        [e-mail]
Subject: Re: (fairly) new 56K manual

And, in case you were wondering, Motorola has an 800 number you can
use for ordering literature: 1 800 521 6274.  Remember, ask for
DSP56000UM/AD (rev 1).  Thanks, Mike!

                        Phil


From @relay.cs.net,@tektronix.tek.com:[e-mail]  Wed Jul 26 23:27:12 1989
Received: from [128.96.32.20] by ka9q.bellcore.com (5.60/4.7)
    id AA15283; Wed, 26 Jul 89 23:27:12 EST
Received: from relay.cs.net by flash.bellcore.com (5.58/1.1)
    id AA26558; Wed, 26 Jul 89 13:00:44 EDT
Received: from tektronix.tek.com by RELAY.CS.NET id aa27152; 26 Jul 89 12:57 EDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA20425; Wed, 26 Jul 89 08:45:52 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA03840; Wed, 26 Jul 89 08:49:47 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA03838; Wed, 26 Jul 89 08:49:42 PDT
Received: by orca.wv.tek.com (5.51/7.1)
    id AA14113; Wed, 26 Jul 89 08:37:00 PDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA18431; Wed, 26 Jul 89 07:33:49 PDT
Received: from tektronix.tek.com by zephyr.ENS.TEK.COM (4.1/7.1)
    id AB19661; Wed, 26 Jul 89 07:31:18 PDT
Received: from rutgers.edu by RELAY.CS.NET id aa07911; 26 Jul 89 9:59 EDT
Received: from mit-eddie.UUCP by rutgers.edu (5.59/SMI4.0/RU1.1/3.04) with UUCP
    id AB11091; Wed, 26 Jul 89 09:57:31 EDT
Received: by EDDIE.MIT.EDU with UUCP with smail2.5 with sendmail-5.45/4.7 id <[e-mail]>; Wed, 26 Jul 89 09:48:31 EDT
Received: by premise.zone1.com (3.2/SMI-3.2)
    id AA29454; Wed, 26 Jul 89 09:03:32 EDT
Date: Wed, 26 Jul 89 09:03:32 EDT
From: Brian Moran <[e-mail]>
Message-Id: <[e-mail]>
To: [e-mail]
Subject: Mundane switched-cap filters...

In the July, 1989 issue of Electronic products, Sierra Semiconductor has
an ad for soem SC22324 "quad biquad filter ic's", which have some EEPROM
on the chip to store the programming params.  Also what is interesting,
is that the stages are programmable serially;  needs of the chip are a
Xtal controlled oscillator, and a programmable divider...  Anyway, is anyone
using this particular chip? I think I'd like to order one, and use it to
build some audio filters that do interesting things -- I can imagine having
one of these in a piece of communication gear with a pretty cheap
microprocessor, and using it to be anything that the operator wanted.

Any thoughts/experiences with this chip?


Brian K. Moran    N9ADG   { ...harvard!mit-eddie,...!mirror}!premise!brian
Premise, Inc.             [e-mail]
3 Cambridge Center
Cambridge, MA 02142       [phone]


From @relay.cs.net,@tektronix.tek.com:[e-mail]  Wed Jul 26 23:28:22 1989
Received: from [128.96.32.20] by ka9q.bellcore.com (5.60/4.7)
    id AA15296; Wed, 26 Jul 89 23:28:22 EST
Received: from relay.cs.net by flash.bellcore.com (5.58/1.1)
    id AA10705; Tue, 25 Jul 89 21:16:04 EDT
Received: from tektronix.tek.com by RELAY.CS.NET id aa23496; 25 Jul 89 21:16 EDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA27918; Tue, 25 Jul 89 17:59:20 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA02690; Tue, 25 Jul 89 18:02:15 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA02687; Tue, 25 Jul 89 18:02:12 PDT
From: "Michael T. Horne" <[e-mail]>
Message-Id: <[e-mail]>
Date: Tue, 25 Jul 89 18:02:11 PDT
To: [e-mail]
Subject: (fairly) new 56K manual
Fcc: outbox

For those of you who aren't aware of this, Motorola has recently printed
a new version of their 56K DSP User's Manual.  It is a more comprehensive
(and coherent) manual on the 56000/56001, including information on their
new revision-C mask that will be out Real Soon Now.  If you'd like to pick
up a copy for your very own, contact your nearest Motorola rep and order
publication number DSP56000UM/AD (rev 1).

Mike



From @relay.cs.net,@tektronix.tek.com:[e-mail]  Wed Jul 26 23:29:29 1989
Received: from [128.96.32.20] by ka9q.bellcore.com (5.60/4.7)
    id AA15306; Wed, 26 Jul 89 23:29:29 EST
Received: from relay.cs.net by flash.bellcore.com (5.58/1.1)
    id AA10718; Tue, 25 Jul 89 21:17:50 EDT
Received: from tektronix.tek.com by RELAY.CS.NET id aa23570; 25 Jul 89 21:17 EDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA27688; Tue, 25 Jul 89 17:49:31 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA02592; Tue, 25 Jul 89 17:53:09 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA02589; Tue, 25 Jul 89 17:53:06 PDT
From: "Michael T. Horne" <[e-mail]>
Message-Id: <[e-mail]>
Date: Tue, 25 Jul 89 17:53:05 PDT
To: [e-mail]
Subject: 12-bit ADCs
Fcc: outbox

I'm curious if anyone on the list has some suggestions on what commercial
12-bit ADCs meet the following constraints:

    a. low cost (less that $20 US, but preferably less than $10)
    b. built-in reference and sample-and-hold
    c. reasonably fast acquisition (>100 kHz)
    c. wide sampling bandwidth
    d. good linearity over the output range
    e. easy (and fast) digital interface
    f. reasonably `blow-up' proof.

I'm most familiar with Analog Devices' parts, as well as some of Maxim's
clones, however there are other manufacturers (e.g. TI, National Semi, TRW,
etc.) that may make parts that meet the above specs.  Any info I can
acquire from the masses saves me time in wading through data books. :)

Price is quite important for my application; anything greater than 20 bucks
is too high.  A built-in reference and SH circuit are simply nice features,
saving external circuitry and reducing overall cost.  A wide sampling
bandwidth (greater than the maximum sampling rate) would be handy for
applications that use undersampling techniques to measure signals above
the theoretical Nyquist limit.

By `easy and fast digital interface' I mean 1) fast data access times (e.g.
Tacc <= 40nS), and 2) reduced access/acquisition synchronization requirements.
Some sort of synchronization between data accesses and initiating a sample
conversion is usually necessary (at least for successive approximation ADCs)
because you can get feedthrough of the digital signal to the converter if
your digital inputs change during the actual conversion process.  Usually the
synchronization is done outside the chip (i.e. by you), however I don't know
why this couldn't be done within the chip.  Any comments?

The last requirement is just for the cases where the analog input greatly
exceeds the input range.  I think most chips clamp above the input range
and are usually good up to the rail voltages, but not above.  Any comments?

If you have a favorite ADC (12-bits or more) that can meet the above specs,
please share it with the group! (are you listening, Bob?:)

Mike



From @relay.cs.net,@tektronix.tek.com:[e-mail]  Wed Jul 26 23:30:39 1989
Received: from [128.96.32.20] by ka9q.bellcore.com (5.60/4.7)
    id AA15321; Wed, 26 Jul 89 23:30:39 EST
Received: from relay.cs.net by flash.bellcore.com (5.58/1.1)
    id AA10233; Tue, 25 Jul 89 20:55:15 EDT
Received: from tektronix.tek.com by RELAY.CS.NET id aa22574; 25 Jul 89 20:54 EDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA27804; Tue, 25 Jul 89 17:54:22 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA02558; Tue, 25 Jul 89 17:31:06 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA02556; Tue, 25 Jul 89 17:31:03 PDT
From: "Michael T. Horne" <[e-mail]>
Message-Id: <[e-mail]>
Date: Tue, 25 Jul 89 17:31:02 PDT
To: [e-mail]
Subject: We have reached steady-state


It would appear that the subscription requests have finally tapered off.
Here is the list in its entirety (this will be the last time I post this).

dsp-group list as of 7/25/89:
------------------------------------------------------------------------------
aa4cg![e-mail]        (Bernie ?)
[e-mail]                (Al Boehnlein)
[e-mail]        (Andy Elms)
[e-mail]                (Alan Biocca - WB6ZQZ)
[e-mail]                (Barry McLarnon - VE3JF)
[e-mail]                (Richard Bisbey - NG6Q)
[e-mail]            (Brian Shiratuki - N6SXT)
[e-mail]                (Bob Albrightson - N7AKR)
[e-mail]            (Maitland Bottoms - AA4HS)
[e-mail]                (Brian K. Moran - N9ADG)
[e-mail]                (Brian Lloyd - WB6RQN)
[e-mail]                (Devon Bowen - KA2NRC)
[e-mail]                    (Bob Buass - K6KGS)
[e-mail]        (Tom Clark - W3IWI)
[e-mail]                (Randy Cole - KN6W)
[e-mail]                (Doug Collinge)
[e-mail]                (Dave Borden - K8MMO)
[e-mail]            (Dave McNeill - WB7BLB)
[e-mail]                    (Don Merritt - K3BGE)
[e-mail]                    (?)
[e-mail]            (DSP archive pseudo-person)
[e-mail]            (Bdale's redist list)
[e-mail]            (David's redist list)
[e-mail]                (John Gilmore - KB6DQC)
[e-mail]        (Dan's redist for DEC)
[e-mail]    (Geert's redist list)
[e-mail]            (Phil's redist list)
[e-mail]    (Ray Gordon - G1XRN)
[e-mail]                    (Doug Faunt - N6TQS)
gatech!stiatl![e-mail]        (John DeArmond - WD4OQC)
[e-mail]        (Garry's redist list)
[e-mail]                (Greg Troxel - N1DAM)
[e-mail]            (Glen R. Sands - KA7AYF)
[e-mail]                (Harris Edelman - KB6OWB)
homxb![e-mail]                (Scott McLellan - ND3P)
[e-mail]                (Club Station)
[e-mail]            (Gary Huntress)
[e-mail]                (Remi Hutin - FE6CNB)
integin![e-mail]            (Stephen Bechtolsheim - KX9G)
ism780c!kodak![e-mail]        (Barry L. Ornitz - WA4VZQ)
[e-mail]            (Jim English - WA7QLB)
[e-mail]            (Jan Steinman - N7JDB)
[e-mail]                    (John B. Milton - N8KSN)
[e-mail]                (James Austin - KA2RVO)
[e-mail]                (Jeffrey Austen - K9JA)
[e-mail]                (John A. Limpert - N3DMC)
[e-mail]            (Kevin Burnett - N6UWS)
[e-mail]            (Jacque Kubley - KA9FJS)
[e-mail]            (Kaj Wiik - OH6EH)
[e-mail]            (Stephen Look - KA9SZW)
[e-mail]                (John Lundin - KA4JSI)
[e-mail]            (Bob McGwier - N4HY)
[e-mail]                (Mike Chepponis - K3MC)
[e-mail]                (Mark Matthews - WA6LZO)
[e-mail]                (Marc ? - KG6KF)
[e-mail]            (Michael Horne - KA7AXD)
microsoft![e-mail]    (Jeff Whitlatch - KO7M)
[e-mail]            (Michael Chace - G6DHU)
mimsy!arinc![e-mail]            (Rick Hambly - WB2TNL)
[e-mail]                (Robert Mullen)
[e-mail]                (Jim Mankin - KB3KJ)
[e-mail]            (?)
[e-mail]                    (Dave Palmer - N6KL)
[e-mail]        (Nicholas Goodall)
[e-mail]                (Dave Trulli - NN2Z)
[e-mail]        (Ofer Lapid - 4X6OJ)
[e-mail]                (Dave Pascoe - KM3T)
[e-mail]            (Pat Davis - KD9UU)
[e-mail]                (Phil Lapsley - N6TCT)
[e-mail]                (Paul Shen)
[e-mail]                (Ron Reisor)
[e-mail]                (Rusty Carruth - N7IKQ)
[e-mail]        (Jarkko Vuori)
sequent!islabs![e-mail]        (John Wipfli)
sequent![e-mail]            (Ken Dove - KF7RZ)
[e-mail]                (Steve H. Goldberg - WQ6L)
[e-mail]            (Stanley Dunn - K3SD)
spox![e-mail]                (Mark Grosen - WB7CFQ)
[e-mail]                (Larry Stewart - K4EO)
[e-mail]            (Mark Lomas)
ucsd.ucsd.edu!telesoft![e-mail]    (Bruce Bergman - N7HAW)
uunet!nikhefh.hep.nl![e-mail]    (Henk Peek - PA0HZP)
[e-mail]            (Dewayne Hendricks - WA8DZP)
[e-mail]            (Walter Doerr)
[e-mail]                (Paul Williamson - KB5MU)
[e-mail]            (William Armitage)
cpoint![e-mail]            (Ken Wolff - K1EA)
[e-mail]            (Dewayne Hendricks - WA8DZP)
------------------------------------------------------------------------------

Now let's here some discussions! :)

Mike



From @relay.cs.net,@tektronix.tek.com:[e-mail]  Tue Aug  8 20:17:13 1989
Received: from flash.bellcore.com by ka9q.bellcore.com (5.60/4.7)
    id AA05179; Tue, 8 Aug 89 20:17:13 EST
Received: from relay.cs.net by flash.bellcore.com (5.58/1.1)
    id AA22965; Tue, 8 Aug 89 21:18:40 EDT
Received: from tektronix.tek.com by RELAY.CS.NET id aa09535; 8 Aug 89 21:02 EDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA20434; Tue, 8 Aug 89 18:02:33 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA01311; Tue, 8 Aug 89 17:43:24 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA01309; Tue, 8 Aug 89 17:43:22 PDT
From: "Michael T. Horne" <[e-mail]>
Message-Id: <[e-mail]>
Date: Tue,  8 Aug 89 17:43:22 PDT
To: [e-mail]
Subject: Forwarded message from David Borden

I recently asked David Borden (K8MMO) about dsp projects currently underway at
AMRAD.  He replied directly to me, but I think it may be of interest to others
on the list.

Here is his message (slightly edited) as forwarded per his request.

Mike

(Dave: could you elaborate on #3 below?  How is the progress?)

------- Forwarded Message

Date: Sun, 30 Jul 89 10:35:19 EDT
>From: David Borden - K8MMO <[e-mail]>
Message-Id: <[e-mail]>
Subject:  A/D Converters
To: [e-mail]

1.  Here in AMRAD, we are experimenting with two A/Ds.  The first is
the Texas Instrument 32041.  This is a 14-bit A/D and D/A.  It hooks
in serial fashion to your DSP chip (we like the Motorola 56001).  It
has filtering also.  On input it has a switched capacitor anti-aliasing
filter and on output it has a low pass reconstruction filter.  They
call this chip the Analog Interface Chip (AIC) and a sample kit is
available from TI for $46.  Lead time is one week.  We may try to build
a IBM PC Clone board using this alone for beginning DSP experimenters.

2.  Our big time A/D converter is the Motorola DSP56ADC16 16-bit Sigma
-Delta technology which also has serial output.  This thing has 96 db
dynamic range and should be good for 100 kHz sample rate or greater.  This
is harder to get.  It is $70 in ceramic only (later they will make plastic
and charge $25 or so).  Motorola has a minimum of $250 for experimental
parts buys, so we bought four.  Their lead time is eight weeks or greater
unless you are a big time customer of theirs.  Then you get samples free
and fast.  We do not know any of these big time customers.

3.  Read the AMRAD Newsletter for our progress using these parts.

------- End of Forwarded Message



From @relay.cs.net,@tektronix.tek.com:[e-mail]  Tue Aug  8 20:32:16 1989
Received: from flash.bellcore.com by ka9q.bellcore.com (5.60/4.7)
    id AA05195; Tue, 8 Aug 89 20:32:16 EST
Received: from relay.cs.net by flash.bellcore.com (5.58/1.1)
    id AA23266; Tue, 8 Aug 89 21:33:45 EDT
Received: from tektronix.tek.com by RELAY.CS.NET id aa09626; 8 Aug 89 21:08 EDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA20681; Tue, 8 Aug 89 18:07:44 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA01329; Tue, 8 Aug 89 17:47:37 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA01327; Tue, 8 Aug 89 17:47:35 PDT
From: "Michael T. Horne" <[e-mail]>
Message-Id: <[e-mail]>
Date: Tue,  8 Aug 89 17:47:35 PDT
To: [e-mail]
Subject: Articles on DSP chips...
Fcc: outbox

FYI

In the August 7 issue of Electronic Engineering Times there are three articles
on DSP chips.  One provides an overview of the evolution of DSP chips and
what is in store for the future.  The second provides some general information
about the differences between the fixed-point parts (e.g. 56K) and the new
floating-point parts (e.g. 96K, DSP32C, etc.).  The last article provides
some insight into the architecture of DSP systems.

The articles are fairly good reading.  They start on page 55...

Mike



From @relay.cs.net,@tektronix.tek.com:[e-mail]  Fri Aug 11 14:38:31 1989
Received: from flash.bellcore.com by ka9q.bellcore.com (5.60/4.7)
    id AA09627; Fri, 11 Aug 89 14:38:31 EST
Received: from relay.cs.net by flash.bellcore.com (5.58/1.1)
    id AA24839; Fri, 11 Aug 89 15:39:47 EDT
Received: from tektronix.tek.com by RELAY.CS.NET id aa04931; 11 Aug 89 13:30 EDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA08636; Fri, 11 Aug 89 10:31:28 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA07796; Fri, 11 Aug 89 10:16:47 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA07794; Fri, 11 Aug 89 10:16:40 PDT
Received: by orca.wv.tek.com (5.51/7.1)
    id AA26753; Fri, 11 Aug 89 10:25:00 PDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA08568; Fri, 11 Aug 89 10:29:52 PDT
Received: from RELAY.CS.NET by zephyr.ENS.TEK.COM (4.1/7.1)
    id AA22720; Fri, 11 Aug 89 10:27:04 PDT
Received: from uunet.uu.net by RELAY.CS.NET id aa21159; 11 Aug 89 13:21 EDT
Received: from hp4nl.nluug.nl by uunet.uu.net (5.61/1.14) with SMTP
    id AA23076; Fri, 11 Aug 89 13:20:25 -0400
Received: from [192.16.183.3] by hp4nl.nluug.nl with SMTP
          id AA18516 (5.58.1.14/2.14); Fri, 11 Aug 89 18:31:57 MET
Received: by nikhefh.hep.nl; Fri, 11 Aug 89 17:51:22 +0200 (MET)
Message-Id: <[e-mail]>
Date: Fri, 11 Aug 89 17:51:22 +0200
From: Henk Peek <[e-mail]>
To: [e-mail]
Subject: Analog Devices DSP Applications Handbooks

Analog devices DSP Applications Handbooks.


I got from Analog Devices 3 Applications handbooks
for the ADSP-2100 family. Total about 500 pages.

ADSP-2100 Applications Handbook, Volume 1
Topics include arithmetic, filters, FFT's, LPC, modem algorithms.

ADSP-2100 Family Applications Handbook, Volume 2
Topics include graphics, pulse-code modulation, multirate filters, DTMF.

ADSP-2100 Family Applications Handbook, Volume 3
Topics include optimized and 2d FFTs, memory interface, multiprocessing,
host interface, sonar beamforming.

It present a compilation of routines for a variety of common digital
signal processing applications. Each routine is a prefaced by a
discussion of the algorithm and data formats that underline the code.
When you want more to known of the architecture or instruction set of
the ADSP-2100 you must read the ADSP-2100 User's Manual.

There aren't ISDN numbers, prices and we got them gratis.

They run a DSP BBS at (617)-461-4258 (USA) 300, 1200 or 2400 baud,
no parity, 8 bits data, 1 stopbit.

I known of the USA DSP BBS's of Motorola and Texas Instrument.
Are there more DSP applications handbooks??

Henk, PA0HZP   [e-mail]



From @relay.cs.net,@tektronix.tek.com:[e-mail]  Fri Aug 11 14:38:45 1989
Received: from flash.bellcore.com by ka9q.bellcore.com (5.60/4.7)
    id AA09635; Fri, 11 Aug 89 14:38:45 EST
Received: from relay.cs.net by flash.bellcore.com (5.58/1.1)
    id AA24895; Fri, 11 Aug 89 15:40:07 EDT
Received: from tektronix.tek.com by RELAY.CS.NET id aa06061; 11 Aug 89 14:40 EDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA11246; Fri, 11 Aug 89 11:41:57 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA07938; Fri, 11 Aug 89 11:22:25 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA07936; Fri, 11 Aug 89 11:22:22 PDT
Received: by orca.wv.tek.com (5.51/7.1)
    id AA29879; Fri, 11 Aug 89 11:31:03 PDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA11032; Fri, 11 Aug 89 11:36:00 PDT
Received: from tektronix.tek.com by zephyr.ENS.TEK.COM (4.1/7.1)
    id AB23553; Fri, 11 Aug 89 11:33:00 PDT
Received: from uunet.uu.net by RELAY.CS.NET id aa21892; 11 Aug 89 14:27 EDT
Received: from hp4nl.nluug.nl by uunet.uu.net (5.61/1.14) with SMTP
    id AA23173; Fri, 11 Aug 89 13:20:53 -0400
Received: from [192.16.183.3] by hp4nl.nluug.nl with SMTP
          id AA18520 (5.58.1.14/2.14); Fri, 11 Aug 89 18:32:00 MET
Received: by nikhefh.hep.nl; Fri, 11 Aug 89 17:50:23 +0200 (MET)
Message-Id: <[e-mail]>
Date: Fri, 11 Aug 89 17:50:23 +0200
From: Henk Peek <[e-mail]>
To: [e-mail],
        [e-mail],
        [e-mail], [e-mail],
        [e-mail],
        [e-mail],
        [e-mail]

Analog devices DSP Applications Handbooks.


I got from Analog Devices 3 Applications handbooks
for the ADSP-2100 family. Total about 500 pages.

ADSP-2100 Applications Handbook, Volume 1
Topics include arithmetic, filters, FFT's, LPC, modem algorithms.

ADSP-2100 Family Applications Handbook, Volume 2
Topics include graphics, pulse-code modulation, multirate filters, DTMF.

ADSP-2100 Family Applications Handbook, Volume 3
Topics include optimized and 2d FFTs, memory interface, multiprocessing,
host interface, sonar beamforming.

It present a compilation of routines for a variety of common digital
signal processing applications. Each routine is a prefaced by a
discussion of the algorithm and data formats that underline the code.
When you want more to known of the architecture or instruction set of
the ADSP-2100 you must read the ADSP-2100 User's Manual.

There aren't ISDN numbers, prices and we got them gratis.

They run a DSP BBS at (617)-461-4258 (USA) 300, 1200 or 2400 baud,
no parity, 8 bits data, 1 stopbit.

I known of the USA DSP BBS's of Motorola and Texas Instrument.
Are there more DSP applications handbooks??

Henk, PA0HZP   [e-mail]



From @relay.cs.net,@tektronix.tek.com:[e-mail]  Mon Aug 14 16:45:01 1989
Received: from flash.bellcore.com by ka9q.bellcore.com (5.60/4.7)
    id AA13505; Mon, 14 Aug 89 16:45:01 EST
Received: from relay.cs.net by flash.bellcore.com (5.58/1.1)
    id AA29927; Mon, 14 Aug 89 17:46:31 EDT
Received: from tektronix.tek.com by RELAY.CS.NET id aa25218; 14 Aug 89 17:44 EDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA01628; Mon, 14 Aug 89 14:46:37 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA13704; Mon, 14 Aug 89 14:30:58 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA13702; Mon, 14 Aug 89 14:30:56 PDT
From: "Michael T. Horne" <[e-mail]>
Message-Id: <[e-mail]>
Date: Mon, 14 Aug 89 14:30:56 PDT
To: [e-mail]
Cc: [e-mail]
Subject: Fixed-point operations library?
Fcc: outbox

I'm in need of a library of fixed-point math routines that work with 32-bit
ints stored in arbitrary int/fractional notation.  The basics of fixed-point
math are simple (you need only keep track of the decimal point), but it
would be nice if someone has already written a set of routines for doing
the basic operations.  Is anyone aware of such a library?

Mike



From @relay.cs.net,@tektronix.tek.com:[e-mail]  Tue Aug 15 13:05:06 1989
Received: from flash.bellcore.com by ka9q.bellcore.com (5.60/4.7)
    id AA14622; Tue, 15 Aug 89 13:05:06 EST
Received: from relay.cs.net by flash.bellcore.com (5.58/1.1)
    id AA27956; Tue, 15 Aug 89 14:06:42 EDT
Received: from tektronix.tek.com by RELAY.CS.NET id aa06469; 15 Aug 89 14:05 EDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA04212; Tue, 15 Aug 89 11:06:23 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA02351; Tue, 15 Aug 89 10:46:36 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA02349; Tue, 15 Aug 89 10:46:34 PDT
From: "Michael T. Horne" <[e-mail]>
Message-Id: <[e-mail]>
Date: Tue, 15 Aug 89 10:46:33 PDT
To: [e-mail]
Subject: New newsgroup `comp.dsp'

------- Forwarded Message

Path: orca!zephyr.ens.tek.com!uunet!lll-winken!xanth!ukma!rutgers!anise.acc.com!ivucsb!todd
From: [e-mail] (Todd Day)
Newsgroups: news.groups
Subject: OFFICIAL CALL FOR VOTES: comp.dsp
Message-ID: <[e-mail]>
Date: 14 Aug 89 12:27:24 GMT
Reply-To: [e-mail] (Todd Day)
Organization: Disillusioned Graduate Hackers, Santa Barbara, CA
Lines: 41

This is the official call for votes for proposed group comp.dsp.

Please read the official proposal posted in news.groups or
news.announce.newgroups before proceeding.

How to Vote!
------------

Votes will be taken until Wednesday, September 13th, 1989 (30 days
from today).

YES:
    If you agree that the group comp.dsp should be created as
    specified in the official proposal, send mail to

        [e-mail]

NO:
    If you think that the group comp.dsp should not be created
    as proposed, send mail to

        [e-mail]

COMMENTS:
    If you have some comment you'd like to send to the vote taker,
    or if you get reject messages, send mail to

        [e-mail]
            or
        [e-mail]

PLEASE NOTE:
    You needn't send a subject line or any message text to vote.  In
    fact, please avoid sending any comments with your vote, as comments
    WILL be ignored.

--

Todd Day  |  [e-mail]  |  ivucsb![e-mail]
"It's not the heat, it's the stupidity"  ---  Angry Poodle B-B-Q

------- End of Forwarded Message



From @relay.cs.net,@tektronix.tek.com:[e-mail]  Tue Aug 15 13:12:55 1989
Received: from flash.bellcore.com by ka9q.bellcore.com (5.60/4.7)
    id AA14632; Tue, 15 Aug 89 13:12:55 EST
Received: from relay.cs.net by flash.bellcore.com (5.58/1.1)
    id AA28613; Tue, 15 Aug 89 14:14:30 EDT
Received: from tektronix.tek.com by RELAY.CS.NET id aa06627; 15 Aug 89 14:15 EDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA04516; Tue, 15 Aug 89 11:16:21 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA02373; Tue, 15 Aug 89 10:57:38 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA02371; Tue, 15 Aug 89 10:57:37 PDT
Date: Tue, 15 Aug 89 10:57:37 PDT
From: "Michael T. Horne" <[e-mail]>
Message-Id: <[e-mail]>
To: [e-mail]
Subject: Purpose of comp.dsp

Forwarded from Usenet...
------------------------------------------------------------------------------

> Path: orca!zephyr.ens.tek.com!uunet!anise!ivucsb!todd
> From: [e-mail] (Todd Day)
> Newsgroups: news.groups,comp.misc,comp.sys.next
> Subject: FORMAL PROPOSAL FOR NEW GROUP: comp.dsp
> Message-ID: <[e-mail]>
> Date: 6 Aug 89 18:21:41 GMT
> Reply-To: [e-mail] (Todd Day)
> Followup-To: news.groups
> Organization: Disillusioned Graduate Hackers, Santa Barbara, CA
> Lines: 46
> Xref: orca news.groups:11887 comp.misc:7477 comp.sys.next:3053
>
> [Follow-ups directed to news.groups]
>
> Well, I've gathered up all the responses from the past couple of weeks,
> and have merged them into a formal proposal.  In summation, the proposal
> basically says that if DSP is involved anywhere in the discussion, then
> it belongs in comp.dsp.
>
>
> Proposal for New Group: comp.dsp
>
> 1) Discussions of DSP hardware
>     a) single processors
>     b) DSP boards for computers
>     c) new DSP product announcements
>     d) architecture
> 2) Discussions of DSP software
>     a) source code listings for particular chip
>     b) how to use development software on particular chip
>     c) general purpose DSP software on computers
> 3) Discussions of DSP theory
>     a) general algorithms
>     b) other devices that use DSP (CD players, etc.)
> 4) Discussions of DSP applications
>     a) audio
>     b) image
>     c) control
>     d) communications
>     e) speech
>     f) etc.
>
>
> Now, I know this is broad.  If there is anything you can think of that
> might get deposited in comp.dsp and you think it shouldn't, then let me
> know and I'll modify the proposal.  Of course, any adds would be appreciated
> also.
>
> I will collect all comments until Sunday, August 13th.  At that point, I
> will modify the proposal (if needed) and will issue a call for votes.
>
> Thanks for your help.
>
> --
>
> Todd Day  |  [e-mail]  |  ivucsb![e-mail]
> "It's not the heat, it's the stupidity"  ---  Angry Poodle B-B-Q



From @relay.cs.net,@tektronix.tek.com:[e-mail]  Mon Aug 21 10:29:58 1989
Received: from flash.bellcore.com by ka9q.bellcore.com (5.60/4.7)
    id AA22095; Mon, 21 Aug 89 10:29:58 EST
Received: from relay.cs.net by flash.bellcore.com (5.58/1.1)
    id AA15259; Mon, 21 Aug 89 11:33:00 EDT
Received: from tektronix.tek.com by RELAY.CS.NET id aa00725; 21 Aug 89 11:19 EDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA23239; Mon, 21 Aug 89 08:20:57 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA19802; Mon, 21 Aug 89 08:06:11 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA19798; Mon, 21 Aug 89 08:05:54 PDT
From: "Michael T. Horne" <[e-mail]>
Message-Id: <[e-mail]>
Date: Mon, 21 Aug 89 08:05:53 PDT
To: Ed Satterthwaite <[e-mail]>
Cc: [e-mail], [e-mail]
Subject: Re: Alternative hardware??
In-Reply-To: Your message of Sat, 19 Aug 89 10:43:39 PDT.
             <[e-mail]>


I feel I must make a few comments about processors here as well (when in
Rome)...

One may wish to consider some of the newer DSP processors as SIO controllers,
as well as signal processors.  As an example, consider the Motorola 56K.  The
latest parts run with instruction rates of either 97.5 or 75 nS (the faster
part is a little more expensive).  Coupling these rates with the 24-bit
data bus, one can see that it is very possible to achieve data throughputs
reaching 40 megabytes/sec.  Even if you only get 1/4 that rate (which is
very conservative, since the part can execute regular instructions while
performing parallel moves simultaneously), you can still feed a 80 Mbps link
effortlessly.  In addition, the 56K has dual serial ports, internal RAM, and
reasonably sized address spaces.  Best of all, the prices are reasonable now,
and are expected to drop quickly (currently a 97.5 nS part is ~$50 US).

One potential high-speed link controller might consist of a main processor
(such as an 80186, or even another 56K), with one or more plug in boards
consisting of a single 56K, high-speed RAM, and necessary SIO/link control.
Costs therefore increase as a function of the number of links you expect
to support.

As far as development tools, the tools of choice (currently) are those
distributed by Motorola.  They provide a complete assembler/linker/etc
environment, a very nice simulator (that can simulate multiple parts, also),
and an optional C compiler.  One can easily use the assembler and simulator
to do all development on a PC, and there are many boards available for
downloading your code directly to a PC-based development board.  As for
the C compiler, it `works', but any time-critical code will need to be
hand coded.  All of these tools are reasonably priced for the PC, slightly
more expensive for the Mac II, and are also available (for a rather healthy
fee) for the VAX and Sun 3s.

One can quickly surmise that DSP chips can serve many other applications
beyond signal processing!

Mike



From @relay.cs.net,@tektronix.tek.com:[e-mail]  Tue Aug 22 13:03:40 1989
Received: from flash.bellcore.com by ka9q.bellcore.com (5.60/4.7)
    id AA25297; Tue, 22 Aug 89 13:03:40 EST
Received: from relay.cs.net by flash.bellcore.com (5.58/1.1)
    id AA25639; Tue, 22 Aug 89 13:03:11 EDT
Received: from tektronix.tek.com by RELAY.CS.NET id aa26967; 22 Aug 89 12:54 EDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA08080; Tue, 22 Aug 89 09:52:34 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA01433; Tue, 22 Aug 89 09:32:19 PDT
Received: by orca.wv.tek.com (5.51/7.1)
    id AA29050; Tue, 22 Aug 89 08:29:29 PDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA05170; Tue, 22 Aug 89 08:34:50 PDT
Received: from tektronix.tek.com by zephyr.ENS.TEK.COM (4.1/7.1)
    id AB04965; Tue, 22 Aug 89 08:31:46 PDT
Received: from ucsd.edu by RELAY.CS.NET id aa12484; 22 Aug 89 10:39 EDT
Received: by ucsd.edu; id AA07011
    sendmail 5.61/UCSD-1.0
    Tue, 22 Aug 89 06:17:13 -0700
Received: from vax.cs.pitt.edu by ucsd.edu; id AA06645
    sendmail 5.61/UCSD-1.0 via SMTP
    Tue, 22 Aug 89 06:16:43 -0700 for /usr/lib/sendmail -odq -oi -ftcp-group-request tcp-group-list
Received: by vax.cs.pitt.edu (5.54/1.14)
    id AA23792; Tue, 22 Aug 89 09:01:23 EDT
Date: Tue, 22 Aug 1989 9:01:21 EDT
From: Bob Hoffman <[e-mail]>
To: [e-mail]
Subject: Re: Alternative hardware??
Message-Id: <[e-mail]>

The NeXT computer comes with a Motorola 56001 and its development tools.
We have one on loan for a few weeks and I can see now that I'm going to
have great fun with it, and accomplish little useful work elsewhere!

    ---Bob.



From @relay.cs.net,@tektronix.tek.com:[e-mail]  Wed Aug 23 14:31:53 1989
Received: from flash.bellcore.com by ka9q.bellcore.com (5.60/4.7)
    id AA05636; Wed, 23 Aug 89 14:31:53 EST
Received: from relay.cs.net by flash.bellcore.com (5.58/1.1)
    id AA04036; Wed, 23 Aug 89 15:35:04 EDT
Received: from tektronix.tek.com by RELAY.CS.NET id ab03977; 23 Aug 89 14:12 EDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA23790; Wed, 23 Aug 89 11:13:22 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA07729; Wed, 23 Aug 89 10:54:57 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA07468; Wed, 23 Aug 89 10:43:49 PDT
From: Phil Lapsley <[e-mail]>
Message-Id: <[e-mail]>
Date: Wed, 23 Aug 89 10:43:48 PDT
To: [e-mail]
Subject: Qualcomm Direct Digital Synthesizers

(Due to some local network problems, some dsp-group mail messages are barfing
in my mail queue.  I'm currently attempting to resolve the problem, so
please bear with me...

Mike)
-----------------------------------------------------------------------------

It seems I always see ads for Qualcomm Direct Digital Synthesizers (DDSs)
in every issue of RF Design that I pick up, so I finally asked them for
more info.  Here's a summary.  I'm posting it here because I think a few
people might be interested in it.  Apologies if it comes off sounding like
an ad for Qualcomm.  :-)

Direct digital synthesis generates a waveform of a particular frequency
by accumulating phase changes at a rate controlled by a master clock.
The maximum frequency that can be generated is half that of the clock
frequency (really slightly less because of imperfect filters); low-pass
filtering the result gives you a sine wave.

The Q2334 DDS is a dual direct digital synthesizer in a 68 pin
PLCC that draws about 800 mW off of a single supply.  There are two
separate DDSs on the chip, both of which are accessed through registers.
Each synthesizer has two 32 bit phase increment registers that control its
frequency.  Besides just generating sine waves, the chip can do FSK, PSK,
and MSK:

    FSK    Load the phase increments for the two frequencies into
        the two phase increment registers, and switch between
        them to switch frequencies.

    PSK    Two ways: use the three bits of "external phase modification"
        per synthesizer to modify the phase output in 45 degree
        increments.  These are external pins that allow for easy
        interface to the outside world.  Or,

        Use the upper 8 bits of the second phase increment register
        for a given synthesizer to modify the phase output by 360/256
        (1.4) degrees.

    MSK    One increment register is loaded with an incremental value,
        and the other with a "decremental" value.

The phase can be changed asynchronously or synchronously with an external
clock.  Oh, yeah: it outputs 12 bits per synthesizer that must then be fed
into a DAC and a low pass filter.

They offer 20 MHz, 30 MHz, and 50 MHz parts.  In quantity 2-99 (they
claim minimum order is 2), they are $85, $149, and $169 each.  They also
offer an evaluation kit that contains a 30 MHz Q2334, a 10 bit DAC, and a
RS 232 interface.  Bad news is it's $545 in kit form.

For more information, call Qualcomm at [phone].  The data sheet
is interesting reading.  (I was going to say "good reading", but then I
remembered we're talking about a data sheet here.  :-)

--

I think you could do a nifty QRP rig with one of these!  And when they
come out with a 60 MHz version, we'll have full HF coverage.

                        Phil



From @relay.cs.net,@tektronix.tek.com:[e-mail]  Wed Aug 23 14:31:55 1989
Received: from flash.bellcore.com by ka9q.bellcore.com (5.60/4.7)
    id AA05637; Wed, 23 Aug 89 14:31:55 EST
Received: from relay.cs.net by flash.bellcore.com (5.58/1.1)
    id AA04045; Wed, 23 Aug 89 15:35:09 EDT
Received: from tektronix.tek.com by RELAY.CS.NET id aa04688; 23 Aug 89 14:26 EDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA24216; Wed, 23 Aug 89 11:28:11 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA07950; Wed, 23 Aug 89 11:09:54 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA07946; Wed, 23 Aug 89 11:09:17 PDT
From: "Michael T. Horne" <[e-mail]>
Message-Id: <[e-mail]>
Date: Wed, 23 Aug 89 11:09:16 PDT
To: Phil Lapsley <[e-mail]>
Cc: [e-mail]
Subject: Re: Qualcomm Direct Digital Synthesizers
In-Reply-To: Your message of Wed, 23 Aug 89 10:43:48 PDT.
             <[e-mail]>



>The phase can be changed asynchronously or synchronously with an external
>clock.  Oh, yeah: it outputs 12 bits per synthesizer that must then be fed
>into a DAC and a low pass filter....  They offer 20 MHz, 30 MHz, and 50 MHz
> parts...

Don't forget that 12-bit fast DACs aren't the same price as a DAC08! :)
I know there are plenty of 8-bit video DACs out there that could handle
the 50 MHz rate, and they are reasonably priced, but does anyone know of
some fast, 12-bit DACs with clean output (i.e. no/little glitching when
many bits change) that you don't have to pay a fortune for?

Mike



From @relay.cs.net,@tektronix.tek.com:[e-mail]  Wed Aug 23 20:19:39 1989
Received: from flash.bellcore.com by ka9q.bellcore.com (5.60/4.7)
    id AA05966; Wed, 23 Aug 89 20:19:39 EST
Received: from relay.cs.net by flash.bellcore.com (5.58/1.1)
    id AA22056; Wed, 23 Aug 89 21:22:54 EDT
Received: from tektronix.tek.com by RELAY.CS.NET id aa17815; 23 Aug 89 19:57 EDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA06504; Wed, 23 Aug 89 16:58:26 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA00219; Wed, 23 Aug 89 16:41:28 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA00216; Wed, 23 Aug 89 16:41:24 PDT
Received: by orca.wv.tek.com (5.51/7.1)
    id AA20955; Wed, 23 Aug 89 13:55:47 PDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA00131; Wed, 23 Aug 89 14:01:07 PDT
Received: from tektronix.tek.com by zephyr.ENS.TEK.COM (4.1/7.1)
    id AB28618; Wed, 23 Aug 89 13:58:04 PDT
Received: from ka9q.bellcore.com by RELAY.CS.NET id aa29560; 23 Aug 89 14:18 EDT
Received: by ka9q.bellcore.com (5.60/4.7)
    id AA05555; Wed, 23 Aug 89 13:14:20 EST
Date: Wed, 23 Aug 89 13:14:20 EST
From: Mike Horne <[e-mail]>
Message-Id: <[e-mail]>
To: [e-mail]
Subject: Testing.


This is just a test message.  Please ignore.

Mike


From @relay.cs.net,@tektronix.tek.com:[e-mail]  Fri Sep  8 20:22:56 1989
Received: from flash.bellcore.com by ka9q.bellcore.com (5.60/4.7)
    id AA26937; Fri, 8 Sep 89 20:22:56 EST
Received: from relay.cs.net by flash.bellcore.com (5.58/1.1)
    id AA09829; Fri, 8 Sep 89 21:26:27 EDT
Received: from tektronix.tek.com by RELAY.CS.NET id aa22649; 8 Sep 89 20:17 EDT
Received: by tektronix.TEK.COM (5.51/7.1)
    id AA18011; Fri, 8 Sep 89 18:11:42 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA15791; Fri, 8 Sep 89 18:02:05 PDT
Received: by ka7axd.WV.TEK.COM (5.17/6.24)
    id AA15789; Fri, 8 Sep 89 18:02:04 PDT
From: "Michael T. Horne" <[e-mail]>
Message-Id: <[e-mail]>
Date: Fri,  8 Sep 89 18:02:03 PDT
To: [e-mail]
Subject: 56K development tools?


Anyone out there currently have/using the 56K assembler/simulator and/or
C compiler for the Mac II?

Mike