Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!usc!ucselx!bionet!agate!ucbvax!dog.ee.lbl.gov!nosc!crash!simpact!jeh
From: jeh@dcs.simpact.com
Newsgroups: sci.electronics
Subject: Re: IR receiver on traffic lights ? (no, it's visible light.)
Message-ID: <1991Apr12.145200.2270@dcs.simpact.com>
Date: 12 Apr 91 21:52:00 GMT
References: <1991Apr11.211957.7309@lynx.CS.ORST.EDU>
Distribution: na
Organization: Simpact Associates, San Diego CA
Lines: 52

In article <1991Apr11.211957.7309@lynx.CS.ORST.EDU>,
 youngqd@jacobs.cs.orst.edu (Dean Youngquist) writes:
> The fire trucks in my town seem to trip traffic lights to green in
> their direction as they are approaching the intersection.  Does anyone
> know how they do this?  I thought they might have a system using
> infra-red light ?  If it is infra-red how can they work in the day-time?
> Doesn't the sun put out infra-red that would interfere ?

Around here the system (called, I think, "Opticom", one of the less-inventive
names I've heard lately) uses visible light.   A large strobe light atop the
fire truck (or paramedic vehicle), running at a fast, but visible, flicker 
rate, does the job.  Yes, the sun and lots of other things put out visible 
light also, but 

	(a) the pickup unit is not just a CdS cell open to the air; it
	includes a telescope-like arrangement of lenses (it actually looks
	like a little telescope, attached to one of the regular R-Y-G traffic
	light assemblies) so that the pickup area is fairly limited, and 

	(b) not very many things (certainly not the sun) put out anything
	that looks like a strobe at 10 or 15 (or whatever) pps.  

It is amazing how well a signal can be picked out of noise, provided you know
the exact characteristics of the signal in question.  In this case, a simple DC
blocking cap will get rid of most of the ambient light, and then it's pretty
easy to filter for a steep-sided narrow pulse waveform at whatever the right 
freq is.  

At night the pickup will "see" 60 Hz hum from the street lighting, but that's
easy to filter out too.  

Consider how bad static from a poorly-wired car can be in an AM radio, and
you'll have some idea how easy it is to "see" a strobe light, even against a
brighter background.  Since spike waveforms have harmonics that (theoretically)
go all the way up the spectrum, they might be able to get away with looking for 
one, or just a few, of the higher-order harmonics. 

The first units installed here (15 years ago or so) had PAR floodlamps
attached, pointing down the street in each direction from which the
intersection could be controlled.  Once the strobe had command of the
intersection, the appropriate lamp would be turned on to let the driver know
that the intersection was indeed controlled and that the green light wouldn't
suddenly turn yellow.  Later units lack this; apparently the system works so
well that they don't need this confirmation. 

Next time you see a fire truck go by, look at the array of flashing lights
and see if there isn't a big (8" diameter reflector or so) strobe among them.  

	--- Jamie Hanrahan (x1116), Simpact Associates, San Diego CA
Chair, VMS Internals Working Group, U.S. DECUS VAX Systems SIG 
Internet:  jeh@dcs.simpact.com, or if that fails, jeh@crash.cts.com
Uucp:  ...{crash,scubed,decwrl}!simpact!jeh