Re: [Phys-l] Detecting ultrasonic ring tones

[John Denker <jsd@av8n.com>]

On 04/22/2008 01:07 PM, Dwight K. Souder wrote:

> I know the students are using those ultrasonic ringtones, the tones
> that kids can hear, but not adults.  A lot of students tell me that
> this is a problem in our school, especially in some classrooms.
>
> Today, I came across a website that I thought might assist us with
> this problem.  It has instructions on how to make an ultrasonic bat
> detector

It's a charming idea ... but this particular solution is not
well suited.  Look at the numbers:

> > For instance, a western pipistrelle bat emits ultrasonic sound in the
> > range of 53 to 91 kHz. If you divide that frequency by 16, the new
> > frequency range is 3.3 to 5.7 kHz, easily within our hearing range.

Kids are not bats.  The "ultrasonic" ringtones cannot be above
20 kHz;  otherwise the kids wouldn't be able to hear them either.
The frequency band of interest is roughly 15 to 20 kHz, probably 
16 to 19 in practice.

Plan B would be to start with the bat detector and modify it,
but that wouldn't work because of insufficient selectivity.
The band of interest is rather narrow, whereas the modified
bat detector would be only broadly tuned; it would be unduly 
sensitive to frequencies outside this band.

If you don't require a portable solution, if I were doing it I
would just use a computer.  Hook up a decent-quality microphone,
run the data through a FFT, pick out the band of interest, and
calculate the power in that band.  
 -- You don't even need to do the inverse FFT if all you want 
  is a light or meter showing the power in that band.
 -- Or if you want to get fancy, just downshift the sound 
  (using the FFT / inverseFFT ... or using a simple single-
  sideband mixer with no FFT) and then you can listen to all 
  the details of the ringtone.

Also note that bats are /loud/ whereas ringtones don't need 
to be particularly loud, and there's a fair amount of background
noise in the band of interest, so you might have trouble detecting
ringtones from a distance, even with good technology.

It's not worth a major effort, because it's a cat-and-mouse
game.  Soon after you catch the first ultrasonic ringtone,
everybody will just switch to "vibrate".  Or (better) switch
to in-the-ear earpieces that nobody else can hear even from
two feet away.



At some point it becomes easier to detect the _RF_ transmitted
by the cellphones.  That should stand out, well above the
background.