Re: Reflected Signal Problem

[Tim Vanderhoek <vanderh@ecf.utoronto.ca>]

On Thu, Nov 25, 1999 at 10:13:43PM -0500, Joyce Poon wrote:
> From what I have been reading on the webpage, there are only certain
> frequencies you can modulate your signal to for the sensors to work
> (modulate = "key a certain frequency" in Tim language).  IR sensors come
> in kits (an IR emitter + a detector&chip thing).  I figure there aren't

Oh dear Joyce.  You've been doing research again.  :)

The thing about random IR interference, however, is that it's
more-or-less random.  I hope...  My hope is that it's possible to
partly handle this type of interference detection in the microcontroller,
but I need to know a lot more about the characteristics of the
interference, first...  This is where experiment and trial & error
will come in handy...

>                                                         Hence, the signal
> frequencies of all the groups may actually be the same. (eg the Sharp
> model -- which seems to be very popular and widely available-- runs at
> 40MHz.)  That's why I am slightly concerned about the reflected signal
> problem.

Hmm...  I see.  My earlier impression had been that we were to devise
this circuitry ourself.  Reflected signals do become a more serious
problem if our frequency is the same as our enemy's...  I guess first
we need to know what the actual ratio of available-freqences to total-
number-of-groups is.

After that, I can't think of anything better than your "bumper-level
detector" scheme.  I'll note two things: 1) they don't really have to
be bumper-level (should be beacon-level), and 2) your suggestion
(see message-id:  Pine.SGI.3.96.991124224315.11207809A-100000@skule.ecf)
that we simply decide a surface is reflective after bumping-into it
several times, is extremely unsatisfactory.  I can outline the
reasons why, if requested, but you may be able to imagine a few of
the reasons...

What is the cost of an IR detector-emittor pair?


> Secondly, I think "interference" does not refer to intentional 
> interference from other groups, but rather from electrical noise/light

Yes, that's my understanding too.

> shielding the motors.  I don't think we have to worry about other groups
> deliberately interfering with beacon signals.  Apparently, this is very

Yes, this is also what Malone indicated (if you were eavesdropping on
him carefully enough).

> hard to accomplish and I heard Malone said (to another group) that no
> groups had ever succeeded in interfering with beacon signals.

Which I see you were, too.  :)


-- 
Signature withheld by request of author.