Shop OBEX P1 Docs P2 Docs Learn Events
Is there such a thing as an acoustic array/matrix ? — Parallax Forums

Is there such a thing as an acoustic array/matrix ?

ArchiverArchiver Posts: 46,084
edited 2003-06-26 19:22 in General Discussion
I was looking over the CMUcamera example, about how it
divides everything into a grid, so it can track the
color blob

Well, can you do this with sound? In a compact way, I don't
want some big honkin thing, just a little-ish device.

I want to build something that can zero in on birds.
And then of course, either take their picture, or
eviscerate them with a laser. -just kidding

Comments

  • ArchiverArchiver Posts: 46,084
    edited 2003-06-21 02:07
    This would require an array of N directional microphones, where N > 1,
    and N is determined by environmental factors and mike
    sensitivity/slectivity. On open prairie or water, 2 mikes or 2 human
    ears (with normal hearing) would do the task. Ask any duck hunter. [noparse]:)[/noparse]

    Dennis

    Original Message
    From: rdenn25 [noparse]/noparse]mailto:[url=http://forums.parallaxinc.com/group/basicstamps/post?postID=CDsISYSq3BjC2hHrUVgs9XVtbcHvsXSCnsEkDyo9cKsZz1yd6P2N9BqlmgBpJqmQMkVtJw]rad0@a...[/url
    Sent: Friday, June 20, 2003 5:02 PM
    To: basicstamps@yahoogroups.com
    Subject: [noparse][[/noparse]basicstamps] Is there such a thing as an acoustic array/matrix
    ?


    I was looking over the CMUcamera example, about how it
    divides everything into a grid, so it can track the
    color blob

    Well, can you do this with sound? In a compact way, I don't want some
    big honkin thing, just a little-ish device.

    I want to build something that can zero in on birds.
    And then of course, either take their picture, or
    eviscerate them with a laser. -just kidding


    To UNSUBSCRIBE, just send mail to:
    basicstamps-unsubscribe@yahoogroups.com
    from the same email address that you subscribed. Text in the Subject
    and Body of the message will be ignored.


    Your use of Yahoo! Groups is subject to
    http://docs.yahoo.com/info/terms/
  • ArchiverArchiver Posts: 46,084
    edited 2003-06-21 03:56
    --- In basicstamps@yahoogroups.com, Dennis O'Leary <doleary@e...>
    wrote:
    > This would require an array of N directional microphones, where N
    > 1,
    > and N is determined by environmental factors and mike
    > sensitivity/slectivity. On open prairie or water, 2 mikes or 2
    human
    > ears (with normal hearing) would do the task. Ask any duck
    hunter. [noparse]:)[/noparse]
    >
    > Dennis
    >

    bear with me. are you saying that I could make an
    acoustic 'pointer/tracker' with two or three microphone?

    are there any examples of this thing for stamps or microcontrollers?
    thanks
  • ArchiverArchiver Posts: 46,084
    edited 2003-06-21 07:48
    Yes. I still have some 2-dim and 3-dim digitizer systems (trade name
    Graf Pen) from the early 1980s. The pen contained a spark source that
    emitted a sharp audible click when a button was pressed, or at a known
    rate in a "free run" mode. The detectors were bar microphones
    surrounding the digitizing area, oriented in either an X-Y, or X-Y-Z
    configuration. Tracking of the stylus pointer was based on acoustic
    delay to the microphone detectors. The 2-dim system is about 36" x 36",
    and the 3 dim system is about 12" cubed.

    It could be done accurately with a fast microcontroller, running at 20
    MHz or more. The expensive part is probably the microphones.

    Dennis

    Original Message
    From: rdenn25 [noparse]/noparse]mailto:[url=http://forums.parallaxinc.com/group/basicstamps/post?postID=yZ25wh9mCoVAopHW0u6ATdxO2wpBH45Irj5skIdFFscQ8jztVq-w2iuxxSXsXhsoldEfRsw]rad0@a...[/url
    Sent: Friday, June 20, 2003 7:56 PM
    To: basicstamps@yahoogroups.com
    Subject: [noparse][[/noparse]basicstamps] Re: Is there such a thing as an acoustic
    array/matrix ?


    --- In basicstamps@yahoogroups.com, Dennis O'Leary <doleary@e...>
    wrote:
    > This would require an array of N directional microphones, where N
    > 1,
    > and N is determined by environmental factors and mike
    > sensitivity/slectivity. On open prairie or water, 2 mikes or 2
    human
    > ears (with normal hearing) would do the task. Ask any duck
    hunter. [noparse]:)[/noparse]
    >
    > Dennis
    >

    bear with me. are you saying that I could make an
    acoustic 'pointer/tracker' with two or three microphone?

    are there any examples of this thing for stamps or microcontrollers?
    thanks



    To UNSUBSCRIBE, just send mail to:
    basicstamps-unsubscribe@yahoogroups.com
    from the same email address that you subscribed. Text in the Subject
    and Body of the message will be ignored.


    Your use of Yahoo! Groups is subject to
    http://docs.yahoo.com/info/terms/
  • ArchiverArchiver Posts: 46,084
    edited 2003-06-23 15:26
    At 12:01 AM 6/21/03 +0000, you wrote:
    >I was looking over the CMUcamera example, about how it
    >divides everything into a grid, so it can track the
    >color blob
    >
    >Well, can you do this with sound? In a compact way, I don't
    >want some big honkin thing, just a little-ish device.
    >
    >I want to build something that can zero in on birds.
    >And then of course, either take their picture, or
    >eviscerate them with a laser. -just kidding

    You can do this sort of thing with triangulation by measuring
    the intensity of three directional microphones arranged 120
    degrees facing away from center. I did this as a kid with
    thunder and lightning to pinpoint direction as well as distance
    of a strike.

    -Beau Schwabe
  • ArchiverArchiver Posts: 46,084
    edited 2003-06-23 22:07
    The goal here would be to measure the phase difference
    (difference in arrival times) at each microphone, then
    swing the two microphones to zero out the phase
    difference -- or use the phase difference to get
    a direction to the source. I'm not sure this
    is a task for the BS2, since in a 'noisy'
    environment determining the identity of a single
    sound, to find the start time of it,
    could be difficult.

    Human hearing is VERY good at differentiating
    sounds, and using the phase difference to
    determine a direction to the sound. That
    doesn't mean it's easy for a CPU.

    The CMU-Cam has an on-board processor which
    processes the pixels for you, to determine
    locations and movement for you.


    --- In basicstamps@yahoogroups.com, "rdenn25" <rad0@a...> wrote:
    > --- In basicstamps@yahoogroups.com, Dennis O'Leary <doleary@e...>
    > wrote:
    > > This would require an array of N directional microphones, where N
    > > 1,
    > > and N is determined by environmental factors and mike
    > > sensitivity/slectivity. On open prairie or water, 2 mikes or 2
    > human
    > > ears (with normal hearing) would do the task. Ask any duck
    > hunter. [noparse]:)[/noparse]
    > >
    > > Dennis
    > >
    >
    > bear with me. are you saying that I could make an
    > acoustic 'pointer/tracker' with two or three microphone?
    >
    > are there any examples of this thing for stamps or microcontrollers?
    > thanks
  • ArchiverArchiver Posts: 46,084
    edited 2003-06-23 22:50
    Yeah OK, I see what you are saying

    instead of a microphone, to hook to the stamp,
    it would take an 'ear' to go to the stamp

    this doesn't exist, does it...

    thanks for the explanation

    Original Message
    From: "Allan Lane" <allan.lane@h...>
    To: <basicstamps@yahoogroups.com>
    Sent: Monday, June 23, 2003 4:07 PM
    Subject: [noparse][[/noparse]basicstamps] Re: Is there such a thing as an acoustic array/matrix
    ?


    > The goal here would be to measure the phase difference
    > (difference in arrival times) at each microphone, then
    > swing the two microphones to zero out the phase
    > difference -- or use the phase difference to get
    > a direction to the source. I'm not sure this
    > is a task for the BS2, since in a 'noisy'
    > environment determining the identity of a single
    > sound, to find the start time of it,
    > could be difficult.
    >
    > Human hearing is VERY good at differentiating
    > sounds, and using the phase difference to
    > determine a direction to the sound. That
    > doesn't mean it's easy for a CPU.
    >
    > The CMU-Cam has an on-board processor which
    > processes the pixels for you, to determine
    > locations and movement for you.
    >
    >
    > --- In basicstamps@yahoogroups.com, "rdenn25" <rad0@a...> wrote:
    > > --- In basicstamps@yahoogroups.com, Dennis O'Leary <doleary@e...>
    > > wrote:
    > > > This would require an array of N directional microphones, where N
    > > > 1,
    > > > and N is determined by environmental factors and mike
    > > > sensitivity/slectivity. On open prairie or water, 2 mikes or 2
    > > human
    > > > ears (with normal hearing) would do the task. Ask any duck
    > > hunter. [noparse]:)[/noparse]
    > > >
    > > > Dennis
    > > >
    > >
    > > bear with me. are you saying that I could make an
    > > acoustic 'pointer/tracker' with two or three microphone?
    > >
    > > are there any examples of this thing for stamps or microcontrollers?
    > > thanks
    >
    >
    > To UNSUBSCRIBE, just send mail to:
    > basicstamps-unsubscribe@yahoogroups.com
    > from the same email address that you subscribed. Text in the Subject and
    Body of the message will be ignored.
    >
    >
    > Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
    >
    >
  • ArchiverArchiver Posts: 46,084
    edited 2003-06-24 02:36
    Do a search on:
    acoustic sonar array
    acoustic sensor array
    phased array microphone.

    For example:
    http://www.jamminpower.com/PDF/Phased-Array%20Microphone.htm


    These things exist . They are used for sonar detection and for acoustic lab
    work, such as automobile noise analysis.

    rad0 <rad0@a...> wrote:
    Yeah OK, I see what you are saying

    instead of a microphone, to hook to the stamp,
    it would take an 'ear' to go to the stamp

    this doesn't exist, does it...

    thanks for the explanation

    Original Message
    From: "Allan Lane"
    To:
    Sent: Monday, June 23, 2003 4:07 PM
    Subject: [noparse][[/noparse]basicstamps] Re: Is there such a thing as an acoustic array/matrix
    ?


    > The goal here would be to measure the phase difference
    > (difference in arrival times) at each microphone, then
    > swing the two microphones to zero out the phase
    > difference -- or use the phase difference to get
    > a direction to the source. I'm not sure this
    > is a task for the BS2, since in a 'noisy'
    > environment determining the identity of a single
    > sound, to find the start time of it,
    > could be difficult.
    >
    > Human hearing is VERY good at differentiating
    > sounds, and using the phase difference to
    > determine a direction to the sound. That
    > doesn't mean it's easy for a CPU.
    >
    > The CMU-Cam has an on-board processor which
    > processes the pixels for you, to determine
    > locations and movement for you.
    >
    >
    > --- In basicstamps@yahoogroups.com, "rdenn25" wrote:
    > > --- In basicstamps@yahoogroups.com, Dennis O'Leary
    > > wrote:
    > > > This would require an array of N directional microphones, where N
    > > > 1,
    > > > and N is determined by environmental factors and mike
    > > > sensitivity/slectivity. On open prairie or water, 2 mikes or 2
    > > human
    > > > ears (with normal hearing) would do the task. Ask any duck
    > > hunter. [noparse]:)[/noparse]
    > > >
    > > > Dennis
    > > >
    > >
    > > bear with me. are you saying that I could make an
    > > acoustic 'pointer/tracker' with two or three microphone?
    > >
    > > are there any examples of this thing for stamps or microcontrollers?
    > > thanks
    >
    >
    > To UNSUBSCRIBE, just send mail to:
    > basicstamps-unsubscribe@yahoogroups.com
    > from the same email address that you subscribed. Text in the Subject and
    Body of the message will be ignored.
    >
    >
    > Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
    >
    >


    To UNSUBSCRIBE, just send mail to:
    basicstamps-unsubscribe@yahoogroups.com
    from the same email address that you subscribed. Text in the Subject and Body of
    the message will be ignored.


    Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/





    Do you Yahoo!?
    SBC Yahoo! DSL - Now only $29.95 per month!

    [noparse][[/noparse]Non-text portions of this message have been removed]
  • ArchiverArchiver Posts: 46,084
    edited 2003-06-24 03:00
    Allan's, Beau's and my comments about sound localization all reduce to:
    you need at least 2 "receivers" (ears, microphones, or whatever), plus
    some electronic processing method of detecting small time differences in
    when the sound arrives at each receiver. Also, note that the
    "processor" must know the location of each of the receivers, and then
    solve some trigonometric equations to localize the sound source. Sounds
    simple, but solving trig equations is not what the Stamp is good at.

    Dennis

    Original Message
    From: rad0 [noparse]/noparse]mailto:[url=http://forums.parallaxinc.com/group/basicstamps/post?postID=4iHAiQl_FJaOmKBd4X1v7khUC9OY5XdaikUuYQMSz3JIGq8tnHjpSx7M9q0Fk7WFNkE4pDpdgGDd]rad0@a...[/url
    Sent: Monday, June 23, 2003 2:50 PM
    To: basicstamps@yahoogroups.com
    Subject: Re: [noparse][[/noparse]basicstamps] Re: Is there such a thing as an acoustic
    array/matrix ?


    Yeah OK, I see what you are saying

    instead of a microphone, to hook to the stamp,
    it would take an 'ear' to go to the stamp

    this doesn't exist, does it...

    thanks for the explanation

    Original Message
    From: "Allan Lane" <allan.lane@h...>
    To: <basicstamps@yahoogroups.com>
    Sent: Monday, June 23, 2003 4:07 PM
    Subject: [noparse][[/noparse]basicstamps] Re: Is there such a thing as an acoustic
    array/matrix ?


    > The goal here would be to measure the phase difference (difference in
    > arrival times) at each microphone, then swing the two microphones to
    > zero out the phase difference -- or use the phase difference to get
    > a direction to the source. I'm not sure this
    > is a task for the BS2, since in a 'noisy'
    > environment determining the identity of a single
    > sound, to find the start time of it,
    > could be difficult.
    >
    > Human hearing is VERY good at differentiating
    > sounds, and using the phase difference to
    > determine a direction to the sound. That
    > doesn't mean it's easy for a CPU.
    >
    > The CMU-Cam has an on-board processor which
    > processes the pixels for you, to determine
    > locations and movement for you.
    >
    >
    > --- In basicstamps@yahoogroups.com, "rdenn25" <rad0@a...> wrote:
    > > --- In basicstamps@yahoogroups.com, Dennis O'Leary <doleary@e...>
    > > wrote:
    > > > This would require an array of N directional microphones, where N
    > > > 1, and N is determined by environmental factors and mike
    > > > sensitivity/slectivity. On open prairie or water, 2 mikes or 2
    > > human
    > > > ears (with normal hearing) would do the task. Ask any duck
    > > hunter. [noparse]:)[/noparse]
    > > >
    > > > Dennis
    > > >
    > >
    > > bear with me. are you saying that I could make an
    > > acoustic 'pointer/tracker' with two or three microphone?
    > >
    > > are there any examples of this thing for stamps or microcontrollers?

    > > thanks
    >
    >
    > To UNSUBSCRIBE, just send mail to:
    > basicstamps-unsubscribe@yahoogroups.com
    > from the same email address that you subscribed. Text in the Subject
    > and
    Body of the message will be ignored.
    >
    >
    > Your use of Yahoo! Groups is subject to
    > http://docs.yahoo.com/info/terms/
    >
    >


    To UNSUBSCRIBE, just send mail to:
    basicstamps-unsubscribe@yahoogroups.com
    from the same email address that you subscribed. Text in the Subject
    and Body of the message will be ignored.


    Your use of Yahoo! Groups is subject to
    http://docs.yahoo.com/info/terms/
  • ArchiverArchiver Posts: 46,084
    edited 2003-06-24 09:32
    thanks gents,

    this is very complicated, I thought that using loudness measurement
    might work, but that's very complicated too
  • ArchiverArchiver Posts: 46,084
    edited 2003-06-24 13:39
    "Doppler Effect" is used to determine distance
  • ArchiverArchiver Posts: 46,084
    edited 2003-06-24 14:03
    Allan,

    A simple approach to this would be to ignore phase and instead use amplitude
    to determine direction. Using two microphones, you would need circuitry to
    amplify (and adjust gain to balance the microphones) and a means to
    determine amplitude of the two microphone signals (possibly a window
    comparator circuit), then, the noise is coming from the general direction of
    the louder signal. Move the microphone pair toward the louder noise and
    when equal that is the direction to the noise source. It is best if the
    noise is filtered to a narrow spectrum for best results as this would tend
    to eliminate some interference from random environmental noise. Phase
    comparison assumes that there are no reflective surfaces near the source or
    receiver (mic) and the characteristics of the noise is known, however, using
    phase along with amplitude can result in excellent performance but with more
    extensive circuitry and signal processing. All the above circuitry can be
    done with op-amps. I know that this involves analog design, but the real
    world is analog, not digital.

    jim
    http://www.geocities.com/jimforkin2003/


    Original Message
    From: Allan Lane [noparse]/noparse]mailto:[url=http://forums.parallaxinc.com/group/basicstamps/post?postID=S87djQ0zEBIpkAQ9Wi2klmLb8le40ZQFSld2-byFvoPsQpU7gIBPyXtqZ7bntOG2AjS5S_xN9gdT8kCFTIlKANBEU4pn]allan.lane@h...[/url
    Sent: Monday, June 23, 2003 5:08 PM
    To: basicstamps@yahoogroups.com
    Subject: [noparse][[/noparse]basicstamps] Re: Is there such a thing as an acoustic
    array/matrix ?


    The goal here would be to measure the phase difference
    (difference in arrival times) at each microphone, then
    swing the two microphones to zero out the phase
    difference -- or use the phase difference to get
    a direction to the source. I'm not sure this
    is a task for the BS2, since in a 'noisy'
    environment determining the identity of a single
    sound, to find the start time of it,
    could be difficult.

    Human hearing is VERY good at differentiating
    sounds, and using the phase difference to
    determine a direction to the sound. That
    doesn't mean it's easy for a CPU.

    The CMU-Cam has an on-board processor which
    processes the pixels for you, to determine
    locations and movement for you.


    --- In basicstamps@yahoogroups.com, "rdenn25" <rad0@a...> wrote:
    > --- In basicstamps@yahoogroups.com, Dennis O'Leary <doleary@e...>
    > wrote:
    > > This would require an array of N directional microphones, where N
    > > 1,
    > > and N is determined by environmental factors and mike
    > > sensitivity/slectivity. On open prairie or water, 2 mikes or 2
    > human
    > > ears (with normal hearing) would do the task. Ask any duck
    > hunter. [noparse]:)[/noparse]
    > >
    > > Dennis
    > >
    >
    > bear with me. are you saying that I could make an
    > acoustic 'pointer/tracker' with two or three microphone?
    >
    > are there any examples of this thing for stamps or microcontrollers?
    > thanks


    To UNSUBSCRIBE, just send mail to:
    basicstamps-unsubscribe@yahoogroups.com
    from the same email address that you subscribed. Text in the Subject and
    Body of the message will be ignored.


    Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
  • ArchiverArchiver Posts: 46,084
    edited 2003-06-24 15:45
    There is a relatively simple method of doing this with three microphones
    as I suggested earlier which involves taking a reading of all three at once
    through a "Peak Detector" circuit (one for each mic), or a "digital snapshot"
    of the ADC output at each mic. I was revisiting this idea last night, knowing
    that the next question on this list would probably be how to interpret the
    three vectors produced from each microphone into one solid vector representing
    the sound origin... following up with this I will post a small visual demo
    program I was playing with last night that lets you control where the source
    is originating and "see" the amplitude each sensor experiences in relation to
    the position of the sound source. This exercise is very intuitive to writing
    STAMP/PIC code to decipher the location of a sound source based on the results
    from your sensors. In a nutshell, it comes down to determining one of three
    quadrants (? tri-rants ?) by a simple process of elimination. The remaining
    two are weighted as a ratio of one another. By doing this you can negate any
    amplitude variations that will most likely be varying at the source (Bird)
    without effecting the direction results.

    -Beau Schwabe






    >Allan's, Beau's and my comments about sound localization all reduce to:
    >you need at least 2 "receivers" (ears, microphones, or whatever), plus
    >some electronic processing method of detecting small time differences in
    >when the sound arrives at each receiver. Also, note that the
    >"processor" must know the location of each of the receivers, and then
    >solve some trigonometric equations to localize the sound source. Sounds
    >simple, but solving trig equations is not what the Stamp is good at.
    >
    >Dennis
    >
    >
    Original Message
    >From: rad0 [noparse]/noparse]mailto:[url=http://forums.parallaxinc.com/group/basicstamps/post?postID=2d9IH5bvmoiZedKZW-qkp3rj05MKze0SCA3OuEpt5mQ5C7TY0WnhrghLs1yIS1bD00EKuQ]rad0@a...[/url
    >Sent: Monday, June 23, 2003 2:50 PM
    >To: basicstamps@yahoogroups.com
    >Subject: Re: [noparse][[/noparse]basicstamps] Re: Is there such a thing as an acoustic
    >array/matrix ?
    >
    >
    >Yeah OK, I see what you are saying
    >
    >instead of a microphone, to hook to the stamp,
    >it would take an 'ear' to go to the stamp
    >
    >this doesn't exist, does it...
    >
    >thanks for the explanation
    >
    >
    Original Message
    >From: "Allan Lane" <allan.lane@h...>
    >To: <basicstamps@yahoogroups.com>
    >Sent: Monday, June 23, 2003 4:07 PM
    >Subject: [noparse][[/noparse]basicstamps] Re: Is there such a thing as an acoustic
    >array/matrix ?
    >
    >
    > > The goal here would be to measure the phase difference (difference in
    > > arrival times) at each microphone, then swing the two microphones to
    > > zero out the phase difference -- or use the phase difference to get
    > > a direction to the source. I'm not sure this
    > > is a task for the BS2, since in a 'noisy'
    > > environment determining the identity of a single
    > > sound, to find the start time of it,
    > > could be difficult.
    > >
    > > Human hearing is VERY good at differentiating
    > > sounds, and using the phase difference to
    > > determine a direction to the sound. That
    > > doesn't mean it's easy for a CPU.
    > >
    > > The CMU-Cam has an on-board processor which
    > > processes the pixels for you, to determine
    > > locations and movement for you.
    > >
    > >
    > > --- In basicstamps@yahoogroups.com, "rdenn25" <rad0@a...> wrote:
    > > > --- In basicstamps@yahoogroups.com, Dennis O'Leary <doleary@e...>
    > > > wrote:
    > > > > This would require an array of N directional microphones, where N
    > > > > 1, and N is determined by environmental factors and mike
    > > > > sensitivity/slectivity. On open prairie or water, 2 mikes or 2
    > > > human
    > > > > ears (with normal hearing) would do the task. Ask any duck
    > > > hunter. [noparse]:)[/noparse]
    > > > >
    > > > > Dennis
    > > > >
    > > >
    > > > bear with me. are you saying that I could make an
    > > > acoustic 'pointer/tracker' with two or three microphone?
    > > >
    > > > are there any examples of this thing for stamps or microcontrollers?
    >
    > > > thanks
    > >
    > >
    > > To UNSUBSCRIBE, just send mail to:
    > > basicstamps-unsubscribe@yahoogroups.com
    > > from the same email address that you subscribed. Text in the Subject
    > > and
    >Body of the message will be ignored.
    > >
    > >
    > > Your use of Yahoo! Groups is subject to
    > > http://docs.yahoo.com/info/terms/
    > >
    > >
    >
    >
    >To UNSUBSCRIBE, just send mail to:
    > basicstamps-unsubscribe@yahoogroups.com
    >from the same email address that you subscribed. Text in the Subject
    >and Body of the message will be ignored.
    >
    >
    >Your use of Yahoo! Groups is subject to
    >http://docs.yahoo.com/info/terms/
    >
    >
    >
    >
    >
    >To UNSUBSCRIBE, just send mail to:
    > basicstamps-unsubscribe@yahoogroups.com
    >from the same email address that you subscribed. Text in the Subject and
    >Body of the message will be ignored.
    >
    >
    >Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
  • ArchiverArchiver Posts: 46,084
    edited 2003-06-24 23:57
    I like the concept of 3 mikes. If you know their fixed locations, say 120
    degrees apart. You should be able to write 3 equations, enter the trig values
    for sines & cosines (angles are fixed) and invert the matrix. Then the problem
    is reduced to multiplications.

    Dennis O'Leary <doleary@e...> wrote:Allan's, Beau's and my comments
    about sound localization all reduce to:
    you need at least 2 "receivers" (ears, microphones, or whatever), plus
    some electronic processing method of detecting small time differences in
    when the sound arrives at each receiver. Also, note that the
    "processor" must know the location of each of the receivers, and then
    solve some trigonometric equations to localize the sound source. Sounds
    simple, but solving trig equations is not what the Stamp is good at.

    Dennis

    Original Message
    From: rad0 [noparse]/noparse]mailto:[url=http://forums.parallaxinc.com/group/basicstamps/post?postID=LjiwhJDkt0_1vR5nJXfFaMeuCPZ-WIlsUY8I0Y65KOhWHd2pR1l487FQyulX3SThPkRfT1L2XQ]rad0@a...[/url
    Sent: Monday, June 23, 2003 2:50 PM
    To: basicstamps@yahoogroups.com
    Subject: Re: [noparse][[/noparse]basicstamps] Re: Is there such a thing as an acoustic
    array/matrix ?


    Yeah OK, I see what you are saying

    instead of a microphone, to hook to the stamp,
    it would take an 'ear' to go to the stamp

    this doesn't exist, does it...

    thanks for the explanation

    Original Message
    From: "Allan Lane"
    To:
    Sent: Monday, June 23, 2003 4:07 PM
    Subject: [noparse][[/noparse]basicstamps] Re: Is there such a thing as an acoustic
    array/matrix ?


    > The goal here would be to measure the phase difference (difference in
    > arrival times) at each microphone, then swing the two microphones to
    > zero out the phase difference -- or use the phase difference to get
    > a direction to the source. I'm not sure this
    > is a task for the BS2, since in a 'noisy'
    > environment determining the identity of a single
    > sound, to find the start time of it,
    > could be difficult.
    >
    > Human hearing is VERY good at differentiating
    > sounds, and using the phase difference to
    > determine a direction to the sound. That
    > doesn't mean it's easy for a CPU.
    >
    > The CMU-Cam has an on-board processor which
    > processes the pixels for you, to determine
    > locations and movement for you.
    >
    >
    > --- In basicstamps@yahoogroups.com, "rdenn25" wrote:
    > > --- In basicstamps@yahoogroups.com, Dennis O'Leary
    > > wrote:
    > > > This would require an array of N directional microphones, where N
    > > > 1, and N is determined by environmental factors and mike
    > > > sensitivity/slectivity. On open prairie or water, 2 mikes or 2
    > > human
    > > > ears (with normal hearing) would do the task. Ask any duck
    > > hunter. [noparse]:)[/noparse]
    > > >
    > > > Dennis
    > > >
    > >
    > > bear with me. are you saying that I could make an
    > > acoustic 'pointer/tracker' with two or three microphone?
    > >
    > > are there any examples of this thing for stamps or microcontrollers?

    > > thanks
    >
    >
    > To UNSUBSCRIBE, just send mail to:
    > basicstamps-unsubscribe@yahoogroups.com
    > from the same email address that you subscribed. Text in the Subject
    > and
    Body of the message will be ignored.
    >
    >
    > Your use of Yahoo! Groups is subject to
    > http://docs.yahoo.com/info/terms/
    >
    >


    To UNSUBSCRIBE, just send mail to:
    basicstamps-unsubscribe@yahoogroups.com
    from the same email address that you subscribed. Text in the Subject
    and Body of the message will be ignored.


    Your use of Yahoo! Groups is subject to
    http://docs.yahoo.com/info/terms/





    To UNSUBSCRIBE, just send mail to:
    basicstamps-unsubscribe@yahoogroups.com
    from the same email address that you subscribed. Text in the Subject and Body of
    the message will be ignored.


    Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/




    Do you Yahoo!?
    SBC Yahoo! DSL - Now only $29.95 per month!

    [noparse][[/noparse]Non-text portions of this message have been removed]
  • ArchiverArchiver Posts: 46,084
    edited 2003-06-25 15:27
    Here is the QBASIC program that I have been twiddling with
    implementing 3 mics arranged 120 deg apart. The only real
    thing to consider mechanically is that the mics are directional.
    A simple solution would be to cut the bottom 1-2 inches off of
    an 8 inch diameter plastic barrel and place your three mics
    about 8 and 3/8 inches apart along the circumference. Or some
    other "circular" form in place of the barrel.

    '
    QBASIC program Starts here

    Pi# = ATN(1) * 4
    '
    Screen Setup
    SCREEN 12
    WINDOW (-266, -200)-(266, 200)
    FOR Deg = 0 TO 360 STEP .5
    Rad# = (Deg / 180) * Pi#
    X1 = COS(Rad#) * 74
    Y1 = SIN(Rad#) * 74
    X2 = COS(Rad#) * 86
    Y2 = SIN(Rad#) * 86
    PSET (X1, Y1), 14
    PSET (X2, Y2), 14
    NEXT Deg
    LOCATE 3, 59: PRINT "A B C Sensor values."
    LOCATE 4, 58: PRINT "-"
    LOCATE 15, 58: PRINT "+"
    LOCATE 2, 1: PRINT "Use '+' and '-' keys to move source"
    LOCATE 3, 1: PRINT "Use 'Esc' key to exit"
    LOCATE 5, 1: COLOR 14: PRINT "Yellow"; : COLOR 15:
    PRINT " bar in the circle is the Signal source."
    LOCATE 6, 1: COLOR 1: PRINT "Blue"; : COLOR 15:
    PRINT " bar in the circle points to the Calculated source."

    '
    Calculate Fixed sensor position and Draw
    Rad1# = (0 / 180) * Pi#
    Rad2# = (120 / 180) * Pi#
    Rad3# = (240 / 180) * Pi#
    Xsensor1 = COS(Rad1#) * 65
    Ysensor1 = SIN(Rad1#) * 65
    Xsensor2 = COS(Rad2#) * 65
    Ysensor2 = SIN(Rad2#) * 65
    Xsensor3 = COS(Rad3#) * 65
    Ysensor3 = SIN(Rad3#) * 65
    CIRCLE (Xsensor1, Ysensor1), 5, 14
    LOCATE 15, 49: PRINT "A"
    CIRCLE (Xsensor2, Ysensor2), 5, 14
    LOCATE 12, 37: PRINT "B"
    CIRCLE (Xsensor3, Ysensor3), 5, 14
    LOCATE 19, 37: PRINT "C"

    '
    Main Program Loop
    Deg = 0
    WHILE Z <> 1
    Z = INP(&H60)
    Z$ = INKEY$
    WAIT 986, 8 'screen refresh delay

    '
    Keypress logic
    IF Z = 78 OR Z = 13 THEN '+' Key
    Deg = Deg + 5
    IF Deg >= 360 THEN Deg = 0
    END IF
    IF Z = 74 OR Z = 12 THEN '-' Key
    Deg = Deg - 5
    IF Deg <= 0 THEN Deg = 360
    END IF

    '
    Calculate Source Location and Draw indicator
    Rad1# = (Deg / 180) * Pi#
    XsourceOld1 = Xsource1
    YsourceOld1 = Ysource1
    XsourceOld2 = Xsource2
    YsourceOld2 = Ysource2
    Xsource1 = COS(Rad1#) * 75
    Ysource1 = SIN(Rad1#) * 75
    Xsource2 = COS(Rad1#) * 85
    Ysource2 = SIN(Rad1#) * 85
    IF XsourceOld1 <> Xsource1 OR YsourceOld1 <> Ysource1 THEN
    LINE (XsourceOld1, YsourceOld1)-(XsourceOld2, YsourceOld2), 0
    LINE (Xsource1, Ysource1)-(Xsource2, Ysource2), 14
    END IF

    '
    Calculate Each Sensor Distance from Source Location
    DX1Old = DX1
    DX2Old = DX2
    DX3Old = DX3
    DX1 = INT(SQR((Xsource1 - Xsensor1) ^ 2 + (Ysource1 - Ysensor1) ^ 2))
    DX2 = INT(SQR((Xsource1 - Xsensor2) ^ 2 + (Ysource1 - Ysensor2) ^ 2))
    DX3 = INT(SQR((Xsource1 - Xsensor3) ^ 2 + (Ysource1 - Ysensor3) ^ 2))


    'Note there is an inversion that takes place here between the relationship
    'of equating Distance to Amplitude that should be considered.

    '---Determine Sensor that is furthest away ...Notice >= positioning...
    IF DX1 >= DX2 AND DX1 > DX3 THEN
    C1 = 12
    Position = 1 + (DX2 / (DX2 + DX3))
    ELSE
    C1 = 10
    END IF
    IF DX2 > DX1 AND DX2 >= DX3 THEN
    C2 = 12
    Position = 2 + (DX3 / (DX1 + DX3))
    ELSE
    C2 = 10
    END IF
    IF DX3 >= DX1 AND DX3 > DX2 THEN
    C3 = 12
    Position = 0 + (DX1 / (DX1 + DX2))
    ELSE
    C3 = 10
    END IF

    '---Calculate Source direction from the two closest sensors and display---
    '
    'In this example the Calculated value is in Radians...
    '
    '...if you want Degrees:
    '
    ' DEGPosition = (Position * 360) / 3
    '
    '...if you want a value ranging from 0 to 100 then...
    '
    '
    ' Percent = (Position * 100) / 3
    '
    OldXPos = XPosition
    OldYPos = YPosition
    RADPosition# = (Position * Pi# * 2) / 3
    XPosition = COS(RADPosition#) * 40
    YPosition = SIN(RADPosition#) * 40
    IF OldXPos <> XPosition OR OldYPos <> YPosition THEN
    LINE (OldXPos, OldYPos)-(0, 0), 0
    ELSE
    LINE (XPosition, YPosition)-(0, 0), 1
    END IF

    '
    Visually indicate Furthest sensor
    PAINT (Xsensor1, Ysensor1), C1, 14
    PAINT (Xsensor2, Ysensor2), C2, 14
    PAINT (Xsensor3, Ysensor3), C3, 14

    '
    Visually indicate each sensors strength
    IF DX1Old <> DX1 THEN
    LINE (120, 150)-(125, DX1Old), C1, BF
    LINE (120, 0)-(125, DX1), 0, BF
    END IF
    IF DX2Old <> DX2 THEN
    LINE (134, 150)-(139, DX2Old), C2, BF
    LINE (134, 0)-(139, DX2), 0, BF
    END IF
    IF DX3Old <> DX3 THEN
    LINE (148, 150)-(153, DX3Old), C3, BF
    LINE (148, 0)-(153, DX3), 0, BF
    END IF
    WEND
    '
    QBASIC program Ends here
  • ArchiverArchiver Posts: 46,084
    edited 2003-06-25 18:14
    A clever program, Beau. With only one SIN table, and deriving COS from
    it, it might be done with a Stamp if you skip the graphics.
    Dennis

    Original Message
    From: Beau Schwabe [noparse]/noparse]mailto:[url=http://forums.parallaxinc.com/group/basicstamps/post?postID=r_6WWa__0sKqQ6m8ZnzT4-WcEPOplGQeZbR5mYhLN7UkQ2p8MihtJLY2fUwbsNJUmlheMxdjx64MNx26v4VvF_Kl]bschwabe@a...[/url
    Sent: Wednesday, June 25, 2003 7:28 AM
    To: basicstamps@yahoogroups.com
    Subject: RE: [noparse][[/noparse]basicstamps] Re: Is there such a thing as an acoustic
    array/matrix ?


    Here is the QBASIC program that I have been twiddling with implementing
    3 mics arranged 120 deg apart. The only real thing to consider
    mechanically is that the mics are directional. A simple solution would
    be to cut the bottom 1-2 inches off of an 8 inch diameter plastic barrel
    and place your three mics about 8 and 3/8 inches apart along the
    circumference. Or some other "circular" form in place of the barrel.

    '
    QBASIC program Starts here

    Pi# = ATN(1) * 4
    '
    Screen Setup
    SCREEN 12
    WINDOW (-266, -200)-(266, 200) FOR Deg = 0 TO 360 STEP .5
    Rad# = (Deg / 180) * Pi#
    X1 = COS(Rad#) * 74
    Y1 = SIN(Rad#) * 74
    X2 = COS(Rad#) * 86
    Y2 = SIN(Rad#) * 86
    PSET (X1, Y1), 14
    PSET (X2, Y2), 14
    NEXT Deg
    LOCATE 3, 59: PRINT "A B C Sensor values."
    LOCATE 4, 58: PRINT "-"
    LOCATE 15, 58: PRINT "+"
    LOCATE 2, 1: PRINT "Use '+' and '-' keys to move source"
    LOCATE 3, 1: PRINT "Use 'Esc' key to exit"
    LOCATE 5, 1: COLOR 14: PRINT "Yellow"; : COLOR 15:
    PRINT " bar in the circle is the Signal source."
    LOCATE 6, 1: COLOR 1: PRINT "Blue"; : COLOR 15:
    PRINT " bar in the circle points to the Calculated source."

    '
    Calculate Fixed sensor position and Draw
    Rad1# = (0
    / 180) * Pi# Rad2# = (120 / 180) * Pi# Rad3# = (240 / 180) * Pi#
    Xsensor1 = COS(Rad1#) * 65 Ysensor1 = SIN(Rad1#) * 65 Xsensor2 =
    COS(Rad2#) * 65 Ysensor2 = SIN(Rad2#) * 65 Xsensor3 = COS(Rad3#) * 65
    Ysensor3 = SIN(Rad3#) * 65 CIRCLE (Xsensor1, Ysensor1), 5, 14 LOCATE 15,
    49: PRINT "A" CIRCLE (Xsensor2, Ysensor2), 5, 14 LOCATE 12, 37: PRINT
    "B" CIRCLE (Xsensor3, Ysensor3), 5, 14 LOCATE 19, 37: PRINT "C"

    '
    Main Program Loop
    Deg = 0
    WHILE Z <> 1
    Z = INP(&H60)
    Z$ = INKEY$
    WAIT 986, 8 'screen refresh delay

    '
    Keypress logic
    IF Z = 78 OR Z = 13 THEN '+' Key
    Deg = Deg + 5
    IF Deg >= 360 THEN Deg = 0
    END IF
    IF Z = 74 OR Z = 12 THEN '-' Key
    Deg = Deg - 5
    IF Deg <= 0 THEN Deg = 360
    END IF

    '
    Calculate Source Location and Draw indicator
    Rad1# = (Deg / 180) * Pi#
    XsourceOld1 = Xsource1
    YsourceOld1 = Ysource1
    XsourceOld2 = Xsource2
    YsourceOld2 = Ysource2
    Xsource1 = COS(Rad1#) * 75
    Ysource1 = SIN(Rad1#) * 75
    Xsource2 = COS(Rad1#) * 85
    Ysource2 = SIN(Rad1#) * 85
    IF XsourceOld1 <> Xsource1 OR YsourceOld1 <> Ysource1 THEN
    LINE (XsourceOld1, YsourceOld1)-(XsourceOld2, YsourceOld2), 0
    LINE (Xsource1, Ysource1)-(Xsource2, Ysource2), 14
    END IF

    '
    Calculate Each Sensor Distance from Source Location
    DX1Old = DX1
    DX2Old = DX2
    DX3Old = DX3
    DX1 = INT(SQR((Xsource1 - Xsensor1) ^ 2 + (Ysource1 - Ysensor1) ^ 2))
    DX2 = INT(SQR((Xsource1 - Xsensor2) ^ 2 + (Ysource1 - Ysensor2) ^ 2))
    DX3 = INT(SQR((Xsource1 - Xsensor3) ^ 2 + (Ysource1 - Ysensor3) ^ 2))


    'Note there is an inversion that takes place here between the
    relationship 'of equating Distance to Amplitude that should be
    considered.

    '---Determine Sensor that is furthest away ...Notice >= positioning...
    IF DX1 >= DX2 AND DX1 > DX3 THEN
    C1 = 12
    Position = 1 + (DX2 / (DX2 + DX3))
    ELSE
    C1 = 10
    END IF
    IF DX2 > DX1 AND DX2 >= DX3 THEN
    C2 = 12
    Position = 2 + (DX3 / (DX1 + DX3))
    ELSE
    C2 = 10
    END IF
    IF DX3 >= DX1 AND DX3 > DX2 THEN
    C3 = 12
    Position = 0 + (DX1 / (DX1 + DX2))
    ELSE
    C3 = 10
    END IF

    '---Calculate Source direction from the two closest sensors and
    display--- ' 'In this example the Calculated value is in Radians... '
    '...if you want Degrees: '
    ' DEGPosition = (Position * 360) / 3
    '
    '...if you want a value ranging from 0 to 100 then...
    '
    '
    ' Percent = (Position * 100) / 3
    '
    OldXPos = XPosition
    OldYPos = YPosition
    RADPosition# = (Position * Pi# * 2) / 3
    XPosition = COS(RADPosition#) * 40
    YPosition = SIN(RADPosition#) * 40
    IF OldXPos <> XPosition OR OldYPos <> YPosition THEN
    LINE (OldXPos, OldYPos)-(0, 0), 0
    ELSE
    LINE (XPosition, YPosition)-(0, 0), 1
    END IF

    '
    Visually indicate Furthest sensor
    PAINT (Xsensor1, Ysensor1), C1, 14
    PAINT (Xsensor2, Ysensor2), C2, 14
    PAINT (Xsensor3, Ysensor3), C3, 14

    '
    Visually indicate each sensors strength
    IF DX1Old <> DX1 THEN
    LINE (120, 150)-(125, DX1Old), C1, BF
    LINE (120, 0)-(125, DX1), 0, BF
    END IF
    IF DX2Old <> DX2 THEN
    LINE (134, 150)-(139, DX2Old), C2, BF
    LINE (134, 0)-(139, DX2), 0, BF
    END IF
    IF DX3Old <> DX3 THEN
    LINE (148, 150)-(153, DX3Old), C3, BF
    LINE (148, 0)-(153, DX3), 0, BF
    END IF
    WEND
    '
    QBASIC program Ends here









    To UNSUBSCRIBE, just send mail to:
    basicstamps-unsubscribe@yahoogroups.com
    from the same email address that you subscribed. Text in the Subject
    and Body of the message will be ignored.


    Your use of Yahoo! Groups is subject to
    http://docs.yahoo.com/info/terms/
  • ArchiverArchiver Posts: 46,084
    edited 2003-06-25 18:36
    I wonder how to deal with echoes? I haven't done anything this
    sophisticated, but I'm looking at the output of a pulsed ultrasound
    emitter, as received by a mike attached to a 'scope input. The space
    is a room with sheet rock walls and lots of other reflecting
    surfaces. The amplifier gain is turned way up. The ultrasound is so
    directional that often the echo is much stronger than the direct beam.

    -- Tracy


    >Here is the QBASIC program that I have been twiddling with
    >implementing 3 mics arranged 120 deg apart. The only real
    >thing to consider mechanically is that the mics are directional.
    >A simple solution would be to cut the bottom 1-2 inches off of
    >an 8 inch diameter plastic barrel and place your three mics
    >about 8 and 3/8 inches apart along the circumference. Or some
    >other "circular" form in place of the barrel.
    >...
  • ArchiverArchiver Posts: 46,084
    edited 2003-06-25 21:24
    Ultrasound is a different story... It makes just about everything
    reflect sound as though you are in a concrete box.

    Let me re-think this...

    >I wonder how to deal with echoes? I haven't done anything this
    >sophisticated, but I'm looking at the output of a pulsed ultrasound
    >emitter, as received by a mike attached to a 'scope input. The space
    >is a room with sheet rock walls and lots of other reflecting
    >surfaces. The amplifier gain is turned way up. The ultrasound is so
    >directional that often the echo is much stronger than the direct beam.
    >
    > -- Tracy
    >
    >
    > >Here is the QBASIC program that I have been twiddling with
    > >implementing 3 mics arranged 120 deg apart. The only real
    > >thing to consider mechanically is that the mics are directional.
    > >A simple solution would be to cut the bottom 1-2 inches off of
    > >an 8 inch diameter plastic barrel and place your three mics
    > >about 8 and 3/8 inches apart along the circumference. Or some
    > >other "circular" form in place of the barrel.
    > >...
  • ArchiverArchiver Posts: 46,084
    edited 2003-06-25 22:48
    Hi Beau,

    The original post (not from me) was asking about bird calls.
    Definitely not ultrasound. I'm interested in bats. Of course, bats
    are quite good at the echolocation based on two ears. I'm even more
    amazed when I look at ultrasound echos on a 'scope. Must be some
    good "neurodigital" processing going on. Right now we have
    microphones that simply tell if they are there or not. The mike is
    placed low and looking up over a water surface. People always ask
    about automatic localization.

    Here is a funny thing that happened to one of the microphones:
    <http://www.emesystems.com/bathat.htm>

    -- Tracy



    >Ultrasound is a different story... It makes just about everything
    >reflect sound as though you are in a concrete box.
    >
    >Let me re-think this...
    >
    >>I wonder how to deal with echoes? I haven't done anything this
    >>sophisticated, but I'm looking at the output of a pulsed ultrasound
    >>emitter, as received by a mike attached to a 'scope input. The space
    >>is a room with sheet rock walls and lots of other reflecting
    >>surfaces. The amplifier gain is turned way up. The ultrasound is so
    >>directional that often the echo is much stronger than the direct beam.
    >>
    >> -- Tracy
    >>
    >>
    >> >Here is the QBASIC program that I have been twiddling with
    >> >implementing 3 mics arranged 120 deg apart. The only real
    >> >thing to consider mechanically is that the mics are directional.
    >> >A simple solution would be to cut the bottom 1-2 inches off of
    >> >an 8 inch diameter plastic barrel and place your three mics
    >> >about 8 and 3/8 inches apart along the circumference. Or some
    >> >other "circular" form in place of the barrel.
    >> >...
    >
    >
    >
    >To UNSUBSCRIBE, just send mail to:
    > basicstamps-unsubscribe@yahoogroups.com
    >from the same email address that you subscribed. Text in the
    >Subject and Body of the message will be ignored.
    >
    >
    >Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
  • ArchiverArchiver Posts: 46,084
    edited 2003-06-26 01:41
    In the audio biz, there's something called a Head Related Transfer Function
    (HRTF) which is a mathematical model of how we hear a 3D world binaurally
    with only two ears. By calculating timing/phase cues, amplitude differences,
    and comb filtering effects (due to the folds inside our ears), we can easily
    localize a full surround field. The room reflections are indeed a part of
    what our brain learns about its immediate environment, making it much
    quicker at locating an object in a room we've acclimated to. It's also why
    we move our heads from side to side to help figure out where a sound is
    coming from. We're scanning the room and building a model of it in our
    heads.

    So by using a binaural head, something like a plastic dummy head with a pair
    of mic elements in the ear canals, and applying an HRTF function, it should
    be possible to determine the location of a sound source with XYZ
    coordinates.

    Now, this is probably not a simple thing you can do in a stamp. It usually
    taxes the processing power of a souped up P4 or dedicated DSP chip. Still,
    it's something to think about for this sort of thing. There may be a way to
    strip this down to the basic formula if only a single frequency, or perhaps
    multiple tones, were involved.

    Mike Sokol
    www.modernrecording.com
    mikes@m...
    301-739-6842 voice/fax
    301-964-5682 mobile


    Original Message
    From: "Sadler Porter" <porter.sadler@s...>
    To: <basicstamps@yahoogroups.com>
    Sent: Wednesday, June 25, 2003 5:47 PM
    Subject: RE: [noparse][[/noparse]basicstamps] Re: Is there such a thing as an acoustic
    array/matrix ?


    > Yes you have access to very good one. (your ears). Your mind note the
    > slight differences in the time the sound reaches your two ears and that
    give
    > you the since of direction for the source.
    >
    >
    Original Message
    > From: rdenn25 [noparse]/noparse]mailto:[url=http://forums.parallaxinc.com/group/basicstamps/post?postID=NAq61xKKaWH5Y1FImNEh1m6YWTGaq54vCTg2i0ZRik3oQi93CvtBRwK5lwmY9YT6asfEgPx0YZM]rad0@a...[/url
    > Sent: Friday, June 20, 2003 10:56 PM
    > To: basicstamps@yahoogroups.com
    > Subject: [noparse][[/noparse]basicstamps] Re: Is there such a thing as an acoustic
    > array/matrix ?
    >
    >
    >
    > bear with me. are you saying that I could make an
    > acoustic 'pointer/tracker' with two or three microphone?
    >
    > are there any examples of this thing for stamps or microcontrollers?
    > thanks
    >
    >
    >
    > To UNSUBSCRIBE, just send mail to:
    > basicstamps-unsubscribe@yahoogroups.com
    > from the same email address that you subscribed. Text in the Subject and
    > Body of the message will be ignored.
    >
    >
    > Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
    >
    >
    >
    > [noparse][[/noparse]Non-text portions of this message have been removed]
    >
    >
    > To UNSUBSCRIBE, just send mail to:
    > basicstamps-unsubscribe@yahoogroups.com
    > from the same email address that you subscribed. Text in the Subject and
    Body of the message will be ignored.
    >
    >
    > Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
    >
  • ArchiverArchiver Posts: 46,084
    edited 2003-06-26 02:56
    What about a Fourier transform to convert a complex waveform from a
    microphone(s) into discrete frequencies or pigeonholes?
  • ArchiverArchiver Posts: 46,084
    edited 2003-06-26 03:28
    --- In basicstamps@yahoogroups.com, Tracy Allen <tracy@e...> wrote:

    > I wonder how to deal with echoes?
    >
    > -- Tracy
    >
    Could you deal with the echo by using a special horn? A shaped
    inlet for the microphone, simulating the bats' ear lobe. You could
    make a complex shape using fiberglass or composites. I've never
    understood why microphones don't look more like ears.

    The first thing a dog does when he hears something, is reposition
    his ear lobes (horn) and cock his head. The head cock may be more
    communication or emotion than a refinement of his hearing apparatus
    though.
  • ArchiverArchiver Posts: 46,084
    edited 2003-06-26 12:37
    A horn does produce directionality, but doesn't eliminate the room echo
    problem that much. That the ear canal does is cause a frequency dip around 3
    kHz or so for sounds located directly in front... at 0 degree axis. Sound
    sources at 90 and 180 degree axis (directly beaming into your ear canal)
    sound "brighter" due to this effect. The pinnae, ear canal, etc., all
    interact to cause a constantly changing panorama of frequency response,
    phase shifts, and temporal (timing) cues that allow us to localize sources
    quite easily.

    See
    http://human-factors.arc.nasa.gov/ihh/spatial/papers/pdfs_db/AEStutorial10.0
    2_USETHIS.ppt
    for a pretty good paper on the subject.

    Mike Sokol
    www.modernrecording.com
    mikes@m...


    Original Message
    From: "rdenn25" <rad0@a...>
    To: <basicstamps@yahoogroups.com>
    Sent: Wednesday, June 25, 2003 10:28 PM
    Subject: [noparse][[/noparse]basicstamps] Re: Is there such a thing as an acoustic array/matrix
    ?


    > --- In basicstamps@yahoogroups.com, Tracy Allen <tracy@e...> wrote:
    >
    > > I wonder how to deal with echoes?
    > >
    > > -- Tracy
    > >
    > Could you deal with the echo by using a special horn? A shaped
    > inlet for the microphone, simulating the bats' ear lobe. You could
    > make a complex shape using fiberglass or composites. I've never
    > understood why microphones don't look more like ears.
    >
    > The first thing a dog does when he hears something, is reposition
    > his ear lobes (horn) and cock his head. The head cock may be more
    > communication or emotion than a refinement of his hearing apparatus
    > though.
    >
    >
    > To UNSUBSCRIBE, just send mail to:
    > basicstamps-unsubscribe@yahoogroups.com
    > from the same email address that you subscribed. Text in the Subject and
    Body of the message will be ignored.
    >
    >
    > Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
    >
  • ArchiverArchiver Posts: 46,084
    edited 2003-06-26 15:43
    Tracy,

    Yes, I realize the original post was about bird calls...

    Here are just some ideas:

    1)Perhaps we are thinking about the idea of using ultrasound in the
    wrong direction. "Ideally" you would expect an audio signal to decay
    instead of re-enforcing itself. With that said there should be a
    way to de-tune or adjust the sensor position slightly so that the
    signal received does not introduce positive or negative re-enforcement
    based on the surroundings. (precision servo mounted ultrasonic sensors?)
    If your sensors are in a fixed location, then this sort of calibration
    should only need minimal adjustments.

    2)The idea of a special horn might work if the design intention was to
    make the sensor less directional.

    3)Reducing the amount of power driving your ultrasonics might help reduce
    echoing.

    4)What happens if you send a steady bombardment of ultrasound instead
    of pulsing it? - Just curious The "pulsing" could cause extra ringing
    or noise in objects that would otherwise be considered "ultrasonically
    quiet".

    -Beau Schwabe


    >Hi Beau,
    >
    >The original post (not from me) was asking about bird calls.
    >Definitely not ultrasound. I'm interested in bats. Of course, bats
    >are quite good at the echolocation based on two ears. I'm even more
    >amazed when I look at ultrasound echos on a 'scope. Must be some
    >good "neurodigital" processing going on. Right now we have
    >microphones that simply tell if they are there or not. The mike is
    >placed low and looking up over a water surface. People always ask
    >about automatic localization.
    >
    > Here is a funny thing that happened to one of the microphones:
    ><http://www.emesystems.com/bathat.htm>
    >
    > -- Tracy
    >
    >
    >
    > >Ultrasound is a different story... It makes just about everything
    > >reflect sound as though you are in a concrete box.
    > >
    > >Let me re-think this...
    > >
    > >>I wonder how to deal with echoes? I haven't done anything this
    > >>sophisticated, but I'm looking at the output of a pulsed ultrasound
    > >>emitter, as received by a mike attached to a 'scope input. The space
    > >>is a room with sheet rock walls and lots of other reflecting
    > >>surfaces. The amplifier gain is turned way up. The ultrasound is so
    > >>directional that often the echo is much stronger than the direct beam.
    > >>
    > >> -- Tracy
    > >>
    > >>
    > >> >Here is the QBASIC program that I have been twiddling with
    > >> >implementing 3 mics arranged 120 deg apart. The only real
    > >> >thing to consider mechanically is that the mics are directional.
    > >> >A simple solution would be to cut the bottom 1-2 inches off of
    > >> >an 8 inch diameter plastic barrel and place your three mics
    > >> >about 8 and 3/8 inches apart along the circumference. Or some
    > >> >other "circular" form in place of the barrel.
    > >> >...
  • ArchiverArchiver Posts: 46,084
    edited 2003-06-26 16:04
    Another thought...

    At what frequency do bats "PING" their environment?
    Does it vary, if so how much? ...from species to species?

    My thought was, would it be possible to setup a BFO (Beat
    Frequency Oscillator) similar to older style metal detector
    designs... If applied to bats, this would translate into a
    Doppler shift of frequency that would contain information
    relating to the Bats position, stationary or in flight.

    -Beau Schwabe

    >Tracy,
    >
    >Yes, I realize the original post was about bird calls...
    >
    >Here are just some ideas:
    >
    >1)Perhaps we are thinking about the idea of using ultrasound in the
    > wrong direction. "Ideally" you would expect an audio signal to decay
    > instead of re-enforcing itself. With that said there should be a
    > way to de-tune or adjust the sensor position slightly so that the
    > signal received does not introduce positive or negative re-enforcement
    > based on the surroundings. (precision servo mounted ultrasonic sensors?)
    > If your sensors are in a fixed location, then this sort of calibration
    > should only need minimal adjustments.
    >
    >2)The idea of a special horn might work if the design intention was to
    > make the sensor less directional.
    >
    >3)Reducing the amount of power driving your ultrasonics might help reduce
    > echoing.
    >
    >4)What happens if you send a steady bombardment of ultrasound instead
    > of pulsing it? - Just curious The "pulsing" could cause extra ringing
    > or noise in objects that would otherwise be considered "ultrasonically
    > quiet".
    >
    >-Beau Schwabe
    >
    >
    > >Hi Beau,
    > >
    > >The original post (not from me) was asking about bird calls.
    > >Definitely not ultrasound. I'm interested in bats. Of course, bats
    > >are quite good at the echolocation based on two ears. I'm even more
    > >amazed when I look at ultrasound echos on a 'scope. Must be some
    > >good "neurodigital" processing going on. Right now we have
    > >microphones that simply tell if they are there or not. The mike is
    > >placed low and looking up over a water surface. People always ask
    > >about automatic localization.
    > >
    > > Here is a funny thing that happened to one of the microphones:
    > ><http://www.emesystems.com/bathat.htm>
    > >
    > > -- Tracy
    > >
    > >
    > >
    > > >Ultrasound is a different story... It makes just about everything
    > > >reflect sound as though you are in a concrete box.
    > > >
    > > >Let me re-think this...
    > > >
    > > >>I wonder how to deal with echoes? I haven't done anything this
    > > >>sophisticated, but I'm looking at the output of a pulsed ultrasound
    > > >>emitter, as received by a mike attached to a 'scope input. The space
    > > >>is a room with sheet rock walls and lots of other reflecting
    > > >>surfaces. The amplifier gain is turned way up. The ultrasound is so
    > > >>directional that often the echo is much stronger than the direct beam.
    > > >>
    > > >> -- Tracy
    > > >>
    > > >>
    > > >> >Here is the QBASIC program that I have been twiddling with
    > > >> >implementing 3 mics arranged 120 deg apart. The only real
    > > >> >thing to consider mechanically is that the mics are directional.
    > > >> >A simple solution would be to cut the bottom 1-2 inches off of
    > > >> >an 8 inch diameter plastic barrel and place your three mics
    > > >> >about 8 and 3/8 inches apart along the circumference. Or some
    > > >> >other "circular" form in place of the barrel.
    > > >> >...
    >
    >
    >
    >To UNSUBSCRIBE, just send mail to:
    > basicstamps-unsubscribe@yahoogroups.com
    >from the same email address that you subscribed. Text in the Subject and
    >Body of the message will be ignored.
    >
    >
    >Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
  • ArchiverArchiver Posts: 46,084
    edited 2003-06-26 17:37
    Bat spectra vary with species, and are broad-band instead of single
    frequencies. Several models of heterodyne "bat detectors", are
    available at Bat Conservation International. The deluxe model has time
    expansion and a digital display <
    http://www.batcon.org/catalog/batdetectors.html >.

    Dennis

    Original Message
    From: Beau Schwabe [noparse]/noparse]mailto:[url=http://forums.parallaxinc.com/group/basicstamps/post?postID=wd3w07DAg4A_45d9MZb5TzzpmYrEofqZE8IochH0lf_D6VwUW-gH9WHYvaKJX2lWuRLkyIC2H2ctG5-rNYKcAVjB]bschwabe@a...[/url
    Sent: Thursday, June 26, 2003 8:05 AM
    To: basicstamps@yahoogroups.com
    Subject: RE: [noparse][[/noparse]basicstamps] Re: Is there such a thing as an acoustic
    array/matrix ?


    Another thought...

    At what frequency do bats "PING" their environment?
    Does it vary, if so how much? ...from species to species?

    My thought was, would it be possible to setup a BFO (Beat Frequency
    Oscillator) similar to older style metal detector designs... If applied
    to bats, this would translate into a Doppler shift of frequency that
    would contain information relating to the Bats position, stationary or
    in flight.

    -Beau Schwabe

    >Tracy,
    >
    >Yes, I realize the original post was about bird calls...
    >
    >Here are just some ideas:
    >
    >1)Perhaps we are thinking about the idea of using ultrasound in the
    > wrong direction. "Ideally" you would expect an audio signal to
    decay
    > instead of re-enforcing itself. With that said there should be a
    > way to de-tune or adjust the sensor position slightly so that the
    > signal received does not introduce positive or negative
    re-enforcement
    > based on the surroundings. (precision servo mounted ultrasonic
    sensors?)
    > If your sensors are in a fixed location, then this sort of
    calibration
    > should only need minimal adjustments.
    >
    >2)The idea of a special horn might work if the design intention was to
    > make the sensor less directional.
    >
    >3)Reducing the amount of power driving your ultrasonics might help
    reduce
    > echoing.
    >
    >4)What happens if you send a steady bombardment of ultrasound instead
    > of pulsing it? - Just curious The "pulsing" could cause extra
    ringing
    > or noise in objects that would otherwise be considered
    "ultrasonically
    > quiet".
    >
    >-Beau Schwabe
    >
    >
    > >Hi Beau,
    > >
    > >The original post (not from me) was asking about bird calls.
    > >Definitely not ultrasound. I'm interested in bats. Of course, bats
    > >are quite good at the echolocation based on two ears. I'm even more
    > >amazed when I look at ultrasound echos on a 'scope. Must be some
    > >good "neurodigital" processing going on. Right now we have
    > >microphones that simply tell if they are there or not. The mike is
    > >placed low and looking up over a water surface. People always ask
    > >about automatic localization.
    > >
    > > Here is a funny thing that happened to one of the microphones:
    > ><http://www.emesystems.com/bathat.htm>
    > >
    > > -- Tracy
    > >
    > >
    > >
    > > >Ultrasound is a different story... It makes just about everything
    > > >reflect sound as though you are in a concrete box.
    > > >
    > > >Let me re-think this...
    > > >
    > > >>I wonder how to deal with echoes? I haven't done anything this
    > > >>sophisticated, but I'm looking at the output of a pulsed
    > > >>ultrasound emitter, as received by a mike attached to a 'scope
    > > >>input. The space is a room with sheet rock walls and lots of
    > > >>other reflecting surfaces. The amplifier gain is turned way up.
    > > >>The ultrasound is so directional that often the echo is much
    > > >>stronger than the direct beam.
    > > >>
    > > >> -- Tracy
    > > >>
    > > >>
    > > >> >Here is the QBASIC program that I have been twiddling with
    > > >> >implementing 3 mics arranged 120 deg apart. The only real thing

    > > >> >to consider mechanically is that the mics are directional. A
    > > >> >simple solution would be to cut the bottom 1-2 inches off of an
    > > >> >8 inch diameter plastic barrel and place your three mics about 8

    > > >> >and 3/8 inches apart along the circumference. Or some other
    > > >> >"circular" form in place of the barrel. ...
    >
    >
    >
    >To UNSUBSCRIBE, just send mail to:
    > basicstamps-unsubscribe@yahoogroups.com
    >from the same email address that you subscribed. Text in the Subject
    and
    >Body of the message will be ignored.
    >
    >
    >Your use of Yahoo! Groups is subject to
    http://docs.yahoo.com/info/terms/



    To UNSUBSCRIBE, just send mail to:
    basicstamps-unsubscribe@yahoogroups.com
    from the same email address that you subscribed. Text in the Subject
    and Body of the message will be ignored.


    Your use of Yahoo! Groups is subject to
    http://docs.yahoo.com/info/terms/
  • ArchiverArchiver Posts: 46,084
    edited 2003-06-26 19:22
    We are using frequency division detectors made by Titley Electronics
    of Australia. The bat calls can be recorded automatically on a CF
    card. A stamp is monitoring the solar power and environmental
    factors. These frequency division detectors lose the amplitude
    information, but the frequency itself is very telling.

    Bat echolocation signals often sweep rapidly down from a high
    frequency and hover for a few milliseconds at a lower frequency, a
    "J" curve. The main frequency does vary by species and can be from
    18khz to 60 khz depending on species. The signal is labile, even
    from time to time with one bat as it adapts to new conditions. In
    crowded conditions with many bats, the signals can overlap and adapt
    to that too. By analyzing the recorded signals, the researchers can
    tell which bat species were present at what time and also get an idea
    of what they was doing. I think they have even developed some
    computer aided analysis software.

    Right now the mike is place fairly near the ground by the shore, to
    capture signals of both low flying "skimmers" near the shore and also
    high flyers further out. Unfortunately, the budget does not allow a
    lot of duplication.


    -- Tracy




    >Bat spectra vary with species, and are broad-band instead of single
    >frequencies. Several models of heterodyne "bat detectors", are
    >available at Bat Conservation International. The deluxe model has time
    >expansion and a digital display <
    >http://www.batcon.org/catalog/batdetectors.html >.
    >
    >Dennis
    >
    >
    Original Message
    >From: Beau Schwabe [noparse]/noparse]mailto:[url=http://forums.parallaxinc.com/group/basicstamps/post?postID=3xKMvHI3ZjOI0HZXCDK2Igq9HJ4_p-ef6_S0qn310naRgqr2WN5I8b6KzV7tcH46_Xcw96wKTH1SnjThiVuOeJTR]bschwabe@a...[/url
    >Sent: Thursday, June 26, 2003 8:05 AM
    >To: basicstamps@yahoogroups.com
    >Subject: RE: [noparse][[/noparse]basicstamps] Re: Is there such a thing as an acoustic
    >array/matrix ?
    >
    >
    >Another thought...
    >
    >At what frequency do bats "PING" their environment?
    >Does it vary, if so how much? ...from species to species?
    >
    >My thought was, would it be possible to setup a BFO (Beat Frequency
    >Oscillator) similar to older style metal detector designs... If applied
    >to bats, this would translate into a Doppler shift of frequency that
    >would contain information relating to the Bats position, stationary or
    >in flight.
    >
    >-Beau Schwabe
    >
    >>Tracy,
    >>
    >>Yes, I realize the original post was about bird calls...
    >>
    >>Here are just some ideas:
    >>
    >>1)Perhaps we are thinking about the idea of using ultrasound in the
    >> wrong direction. "Ideally" you would expect an audio signal to
    >decay
    >> instead of re-enforcing itself. With that said there should be a
    >> way to de-tune or adjust the sensor position slightly so that the
    >> signal received does not introduce positive or negative
    >re-enforcement
    >> based on the surroundings. (precision servo mounted ultrasonic
    >sensors?)
    >> If your sensors are in a fixed location, then this sort of
    >calibration
    >> should only need minimal adjustments.
    >>
    >>2)The idea of a special horn might work if the design intention was to
    >> make the sensor less directional.
    >>
    >>3)Reducing the amount of power driving your ultrasonics might help
    >reduce
    >> echoing.
    >>
    >>4)What happens if you send a steady bombardment of ultrasound instead
    >> of pulsing it? - Just curious The "pulsing" could cause extra
    >ringing
    >> or noise in objects that would otherwise be considered
    >"ultrasonically
    >> quiet".
    >>
    >>-Beau Schwabe
    > >
    > >
    > > >Hi Beau,
    > > >
    > > >The original post (not from me) was asking about bird calls.
    > > >Definitely not ultrasound. I'm interested in bats. Of course, bats
    > > >are quite good at the echolocation based on two ears. I'm even more
    >> >amazed when I look at ultrasound echos on a 'scope. Must be some
    >> >good "neurodigital" processing going on. Right now we have
    >> >microphones that simply tell if they are there or not. The mike is
    >> >placed low and looking up over a water surface. People always ask
    >> >about automatic localization.
    >> >
    >> > Here is a funny thing that happened to one of the microphones:
    >> ><http://www.emesystems.com/bathat.htm>
    > > >
    >> > -- Tracy
    >> >
    >> >
    >> >
    >> > >Ultrasound is a different story... It makes just about everything
    >> > >reflect sound as though you are in a concrete box.
    >> > >
    >> > >Let me re-think this...
    >> > >
    >> > >>I wonder how to deal with echoes? I haven't done anything this
    >> > >>sophisticated, but I'm looking at the output of a pulsed
    >> > >>ultrasound emitter, as received by a mike attached to a 'scope
    >> > >>input. The space is a room with sheet rock walls and lots of
    >> > >>other reflecting surfaces. The amplifier gain is turned way up.
    >> > >>The ultrasound is so directional that often the echo is much
    >> > >>stronger than the direct beam.
    >> > >>
    >> > >> -- Tracy
    >> > >>
    >> > >>
    >> > >> >Here is the QBASIC program that I have been twiddling with
    >> > >> >implementing 3 mics arranged 120 deg apart. The only real thing
    >
    >> > >> >to consider mechanically is that the mics are directional. A
    >> > >> >simple solution would be to cut the bottom 1-2 inches off of an
    >> > >> >8 inch diameter plastic barrel and place your three mics about 8
    >
    >> > >> >and 3/8 inches apart along the circumference. Or some other
    >> > >> >"circular" form in place of the barrel. ...
    >>
    >>
    >>
    >>To UNSUBSCRIBE, just send mail to:
    >> basicstamps-unsubscribe@yahoogroups.com
    >>from the same email address that you subscribed. Text in the Subject
    >and
    >>Body of the message will be ignored.
    >>
    >>
    >>Your use of Yahoo! Groups is subject to
    >http://docs.yahoo.com/info/terms/
    >
    >
    >
    >To UNSUBSCRIBE, just send mail to:
    > basicstamps-unsubscribe@yahoogroups.com
    >from the same email address that you subscribed. Text in the Subject
    >and Body of the message will be ignored.
    >
    >
    >Your use of Yahoo! Groups is subject to
    >http://docs.yahoo.com/info/terms/
    >
    >
    >
    >
    >
    >To UNSUBSCRIBE, just send mail to:
    > basicstamps-unsubscribe@yahoogroups.com
    >from the same email address that you subscribed. Text in the
    >Subject and Body of the message will be ignored.
    >
    >
    >Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
Sign In or Register to comment.