NetNews Usenet Archive 1993 #3/NN_1993_3.iso/spool/comp/sys/hp48/7219

Internet Message Format | 1993-01-21 | 5.2 KB

Path: sparky!uunet!spool.mu.edu!sgiblab!cs.uoregon.edu!news.uoregon.edu!nntp.uoregon.edu!stevev
From: [e-mail] (Steve VanDevender)
Newsgroups: comp.sys.hp48
Subject: Re: CRAZY_IDEA
Date: 21 Jan 93 00:32:19
Organization: University of Oregon Chemistry Stores
Lines: 89
Message-ID: <[e-mail]>
References: <[e-mail]> <[e-mail]>
    <[e-mail]>
    <[e-mail]>
NNTP-Posting-Host: miser.uoregon.edu
In-reply-to: [e-mail]'s message of 20 Jan 1993 21:00:22 GMT

In article <[e-mail]> [e-mail]
(ochealth) writes:

[about the HP 48 tricorder programs:]

   Perhaps not reliably, but it can detect distance.
   It beeps quickly when something is close.

It is detecting proximity.  It can tell when something is close
enough to return an echo -- but that is not the same as measuring
distance, because my experience has been that the difference
between getting no reflection and continuous reflection is very
small and too dependent on lighting conditions and the surface
that the IR is reflecting off of.

   >The problem is not that the reflected signal might be too weak to
   >measure distances (although that would limit the total range),
   >but that the HP 48 cannot measure time intervals short enough to
   >measure distances less than 300-1000 meters.  Light travels 300
   >meters in 1 microsecond, which is about the smallest time
   >interval you could hope to measure with the HP 48 CPU clock.

   true, but who says you have to measure the time between
   sending the output pulse, and the FIRST one received. If the
   object remains at a fixed distance, a periodic set of IR
   pulses and reflections could be generated.  I don't know how
   accurate this would be, but it reminds me about how you can
   measure a 100 MHz signal, assuming it is a sine wave, but
   without having to sample the signal at 200 MHz. Since you know
   the wave is periodic, you can sample at a much lower rate.

The HP 48 IR detector is digital, not analog.  It cannot tell you
the signal strength.  It can only tell you whether the IR flux
falling on it is above a threshhold or not.  _If_ you have an
ultrastable time base, and _if_ you already know something about
the nature of the high-speed signal you want to sample and _if_
that signal is repeating periodically, and _if_ you can measure
intensity, you might be able to learn something about the
high-frequency signal by sampling it at rate much slower than its
frequency.  However, this dependence on the ultrastability of
your time base is essentially dependence on being able to measure
an exact multiple of some very short time interval.  The
difference between the time interval you wish to measure (on the
order of 1e-11 seconds) and the maximum rate at which you could
toggle the HP 48 IR output (about 1e-5 seconds) is six orders of
magnitude.  The variation in the HP 48's CPU clock is probably
more than one part per million across a time interval of a
second.

   So with the HP, you could send out several pulses, knowing
   that you would miss receiving the first few reflections. If
   you assume the object is a fixed distance away, and you know
   you can't measure better than dT seconds, you can still
   measure the distance by measuring the time delay between
   sending and receiving thousands of pulses. After thousands
   (whatever) of pulses, the time delay will have acucumulated,
   and will be big enough to measure. I don't know if this sounds
   right, but if you think about using a vernier gauge to measure
   small distances, it should be clearer.  (in a vernier, you
   have two sets of markings, all spaced dx apart, but you can
   put the two together, and measure distances smaller than dx)

Think about this.  It takes about 10 microseconds from the time
you could turn on the HP 48 IR transmitter to the time you could
read the IR receiver status.  In that time the IR pulse will
travel 3000 meters.  There is _no way_ you can measure any time
interval (and thus any distance) shorter than this, and the IR
transmitter has been on for that whole 10 microseconds (and can't
be turned off until another 10 microseconds has passed).  The
method you describe assumes that you can generate IR pulses on
the order of 1e-11 seconds in duration, then use an ultrastable
time base to detect an interference pattern in the returned
echoes.  The HP 48 can't generate pulses that short and doesn't
have a time base that stable.  There doesn't seem to be any way
to get around these limitations.

Perhaps it might be possible to detect relative distance in a
small region using measurement of intermittent reflection, but
that isn't accurate enough to be a digital ruler.

Now if the HP 48 had a microphone to go with its beeper, you
might have a better chance of measuring distance using sonar.
The speed of sound is slow enough that the HP 48 CPU clock could
count a few times per each centimeter the sound pulse travels.
--
Steve VanDevender     [e-mail]
"Bipedalism--an unrecognized disease affecting over 99% of the population.
Symptoms include lack of traffic sense, slow rate of travel, and the
classic, easily recognized behavior known as walking."