Connect with us

Ultrasonic PA Beacon

Discussion in 'Electronic Design' started by [email protected], Jun 27, 2005.

Scroll to continue with content
  1. Guest

    We are developing an aerosol sampler that will be worn by students
    while they go thru normal daily activities. The sampler will have
    several collection ports that will be selected depending upon the
    student's location (home, school, other). We need a method of
    identifying the location. GPS has been tried in the past and was found
    to be unreliable in an urban setting. This was replaced by a radio
    beacon with better success (at least at home). The beacon's range at
    school, however, is an issue (we are using the ISM unlicensed band).
    The idea came up about transmitting an ultrasonic "subliminal" signal
    over the school's PA system. Even with high attenuation at 20 KHz there
    may be sufficient signal strength to be detected, especially using
    signal recovery techniques.

    Any suggestions/concerns about implementing this are welcomed.

    Tom
     
  2. John Fields

    John Fields Guest

     
  3. Joerg

    Joerg Guest

    Hello Tom,

    20kHz isn't a great idea. It can be heard by some people and would drive
    them crazy. It would certainly drive most animals crazy. I would never
    do that.

    A low frequency radio system can work. For ideas you could study
    instrument landing systems, especially older versions. For example, how
    an airplane could stay on the glide path in pre-GPS times during a low
    visibility approach.

    There may be a market for this stuff. In our area of El Dorado County in
    California a big natural asbestos scare has started. People in space
    suits milling about and all that. I guess pretty soon they'll start long
    term sampling studies on a larger scale.

    Regards, Joerg
     
  4. Guest

    Thanks Joerg.

    The concept is "inaudible", so low freq is fine (and probably more
    benign). In any event, the actual sound levels would be very weak.

    This particular study is in New York City, and was begun before 9/11
    but gained additional attention due to the pulverised building
    material.

    Tom
     
  5. Joerg

    Joerg Guest

    Hello Tom,
    That can become a challenge in terms of foreign noise. A lot of
    machinery such as air conditioners, fans, pumps, motors and so on
    generate a noise spectrum well into the tens of kHz. Even if you had a
    really clever matched filter algorithm there comes a point when the
    signal is so far down in the noise that positioning would become impossible.

    Also, don't underestimate the ingenuity of students who object to
    wearing one of these. Or studentr who want to foil it just for bragging
    rights. Maybe they bring dog whistles or what not.
    Out here it is tremolite that is found in rock formations. Unfortunately
    people have placed crushed rock onto driveways and roads. In other areas
    smog particles or soot have to be measured.

    Regards, Joerg
     
  6. John Fields

    John Fields Guest

     
  7. There are many options, all depending on your budget. The radio is
    probably your best bet. Since you're in New York, I imagine that your
    school has a high-concentration of always-on computers. You can stick
    a Wi-Fi dongle on each computer, and do the same for the collection
    device (is it custom or off-the-shelf?). The same could be done at
    home.

    Then write a program (Visual Basic or C/C++ or whatever) on the
    portable unit to detect which node it is close to. Note that it is not
    necessary to have customized software on the field nodes - all you need
    to do is induce the stationary nodes to emit an Ethernet frame
    containing their MAC addresses. This means that Wi-Fi access points
    (Linksys/Netgear) will act as field nodes too. Depending on the
    resolution required, you may need to process the RSSI, but probably
    not.

    There is another benefit - you get "free" fixed-points, throughout the
    world. You can build (or find on the web) a database in the lab
    mapping 6-byte MAC address to latitude/longitude coordinates.

    When it comes time to render, you get a much richer visualization from
    your dataset, and get to see dynamic movements over time for the
    samples.

    You can find USB Wi-Fi dongles anywhere, as you know, and if you need
    help structuring the program, drop me a note.

    -Chaud Lapin-
     
  8. Guest

    Thanks for the reply. I will look into whether WiFi is available in the
    schools (probably is).

    Tom
     
  9. Rich Grise

    Rich Grise Guest

    Sure.

    Find trainable monkeys to carry the sniffers around, and train
    them to press the "Library", "Home", "Room 101" and so on,
    buttons when they go into the specific environment. Geez, they
    can train human larvae to do text messaging on a video telephone;
    what should be so hard about training your subjects to push a
    button or three?

    Good Luck!
    Rich
     
  10. Rich Grise

    Rich Grise Guest

    Then it's imperative - train your sniffer carriers.

    If you were installing these devices on, for example, sewer rats,
    then there might be some justification for automatic localization,
    but for heaven's sakes, how cheap are you hiring your help? Are
    they _completely_ illiterate?

    Feh,
    Rich
     
  11. Guest

    I don't what they will pay them, if anything.

    Even the best intentioned - and well paid - workers forget to do
    something important. I will forget to grab something from the garage
    after having scoulded myself repeatedly not to just a couple minutes
    prior.
     
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day

-