I have not used Spread Spectrum, but I have studied it for a few
applications that didn't pan out.

As far as bandwidth efficiency, I can see some improvement, but a great
cost in complexity.  A digital bit stream has a SinX/X bandwidth.  We
allocate bandwidth by frequency so it comes in rectangular blocks.  A
SinX/X curve does not fit well in a rectangular block.  If you allocate
a block big enough to hold the SinX/X then there is significant
under-utilized bandwidth in the allocated block.  DSSS fills a
rectangular bandwidth block better.

Also when there are a few too many users for the available band a Spread
Spectrum system does not fail catastrophically by sitting a new user on
top of an existing user.  Instead all users begin to suffer a rising
noise level and service may not be good, but may be acceptable for the
short term overload.

GPS is a good use for SS.  The satellites don't have to keep track of
which of them is visible where to negotiate channel space.  And you
don't have to give each one its own channel.  If you have so many
satellites visible that they start to interfere with each others
reception then you must have lots of pseudorange data to feed into you
computation.  You will get a good position fix from not-so-good data, if
you have enough data.

{soapbox bold}

What really     raises my hackles is using SS for computer clocks, or clock
modulation to pass emissions tests.  The important point about clock
modulation that most people don't realize is that it does nothing to
reduce the actual interference that customers will see in the field.  It
is a clever scheme to circumvent government regulations and nothing
more.  If a milliwatt of energy at exactly 60MHz will cause a certain
amount of rectification in an audio pre-amp, smearing that same energy
from 59MHz to 61MHZ will cause almost exactly the same disruption.  Few
of the potential victims of EMI have any way to make such fine
distinctions in frequency.  The only real solution to the real problem
is to reduce the energy radiated, or shift it in frequency by a very
large amount, like at least a factor of two or three.

I work with EMI problems all the time in oceanographic equipment,
keeping digital noise out of sensitive sonar front ends.  For me the FCC
regulations are irrelevant.  There is nothing like an inch thick steel
pressure case and a thousand meters of seawater for shielding!  The only
EMI requirements I have to meet are Mother Natures, and she doesn't care
about peaks on a spectrum analyzer.  When a switching supply or a
microprocessor at one end of my instrument is radiating and causing
rectification in a sensitive amp at the other end modulating the
interfering source is not going to help at all.

The FCC regulations are not obstacles to be overcome.  They are tools to
help solve real problems.  When finding ways to comply with the rules,
you need to realize that rules are there to prevent problems
and simply complying to the exact letter of the rule may still leave the
underlying problem unsolved.  The FCC rules sometimes need some common
sense interpretation.  Unless we designers use our common sense, the FCC
will have to rewrite the rules to eliminate loopholes like clock
modulation, which will make the rules more complicated and restrictive
than they currently are.

{/bold /soapbox}

Sherpa Doug

> -----Original Message-----
> From: Sean H. Breheny [mailto:shb7@CORNELL.EDU]
> Sent: Friday, April 12, 2002 2:33 PM
> To: PICLIST@MITVMA.MIT.EDU
> Subject: Re: [OT]:Spread Spectrum Efficiency for Continuous signals
>
>
> Hi Doug,
>
> I'd like to hear more. I have no practical experience with SS, but the
> theoretical arguments that I've heard seem strong enough that
> I believed
> that there was some benefit (especially with regard to more closely
> approaching the theoretical channel capacity). Where does
> reality depart
> significantly from this?
>
> Thanks,
>
> Sean
>
> At 02:07 PM 4/12/02 -0400, you wrote:
> >Spread spectrum is much more hype than real benefit.
> Frequency hopping
> >is good to prevent eavesdropping.  Direct Sequence spreading
> helps you
> >hide in the noise.  Both are useful for military
> applications, but have
> >limited ligitimate civilian use.
> >
> >If you had four channels of space with a varying number of users,
> >sometimes more than four, who couldn't agree on who should use what
> >space when, then spread spectrum helps to arbitrate the use
> of bandwith.
> >  When too many users occupy the band everybody's signal to
> noise ratio
> >degrades, and everybody suffers equally.
> >
> >I can stay on this soapbox for longer than anybody wants to read...
> >
> >Sherpa Doug
> >
> > >
>
> --
> http://www.piclist.com hint: PICList Posts must start with ONE topic:
> [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads
>
>
>

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads