Adaptive Filter | ADPCM | ADSP | ADSP-2181 | Aliasing | AMR | Anti-Aliasing | ARMA | Autocorrelation | AutoCovariance | Beamforming | Bessel | Blackfin | Butterworth | C6713 | CCS | Chebyshev | CIC Filter | Circular Convolution | Code Composer Studio | Comb Filter | Compression | Convolution | Cross Correlation | DCT | Decimation | Deconvolution | Demodulation | DM642 | DSP Boards | DSP/BIOS | DTMF | Echo Cancellation | Equalization | Equalizer | ETSI | EZLITE (Ez-kit Lite) | FFT | FFTW | FIR Filter | Fixed Point | FSK | G.711 | G.723 | G.729 | Gaussian Noise | Goertzel | GPIO | Hilbert Transform | IFFT | IIR Filter | Interpolation | Invariance | JTAG | Kalman | Laplace Transform | Levinson | LPC | McBSP | MIPS | Modulation | MPEG | Multirate | Notch Filter | Nyquist | OFDM | Oversampling | Pink Noise | Pitch | PLL | Polyphase | QAM | QDMA | Quantization | Quantizer | Radar | Random Noise | Reed Solomon | Remez | Resampling | RTDX | Sampling | Sharc | TI C6711 | Undersampling | Viterbi | Wavelets | White Noise | Wiener Filter | Windowing | XDS510PP | Z Transform

Hi all,

First of all, let me state that this is not a joke and I'm perfectly 
serious.

I'm researching design options for a proposed device to facilitate
two-way language-based communication with bottlenose dolphins. One
of the requirements is that this device is wearable by a swimmer in 
the water, and totally stand alone.

As I see it, for the dolphin-to-human side of the system, I have need
of a spacial filtering, target discrimination and tracking subsystem,
in order to get a clean enough signal from a particular dolphin to
make it through a recognition process.

My thought for this is to use a head mounted line array of small
hydrophones and a beam former. But I also wish to seperate the
signals from multiple targets. So I'm thinking that I not only need
several beam formers (one per target), but I also divide up the 
spectrum into many bands and for any given target, reject bands
which do not have a strong signal coming from the same direction
as a particular target. Obviously, I have to aim the beam former based
upon the current target vector as either the dolphin or the swimmer
moves around.

So my question is this: do I put the filter bank before or after the
beam former for a particular target ?

If I filter before, doesn't that mean I need one bank per line array
element per target ?

What DSP chips might be able to handle this problem without needing
Hoover Dam to power them ?

I'm not only an newbie amateur in DSP matters.

Any advice would be appreciated.

Regards,

Michael A. Hobson
warrior (underline) mike (Y2K+1) (at) yahoo (dot) com


Michael A. Hobson
email: thesneakster (at) gmail (dot) com
icq: #2186709
yahoo: warrior_mike2001

______________________________
New DSP Code Snippets Section now Live.   Learn more about the reward program for contributors here.

Michael A. Hobson wrote:
> Hi all,
> 
> First of all, let me state that this is not a joke and I'm perfectly 
> serious.
> 
> I'm researching design options for a proposed device to facilitate
> two-way language-based communication with bottlenose dolphins. One
> of the requirements is that this device is wearable by a swimmer in 
> the water, and totally stand alone.
> 
> As I see it, for the dolphin-to-human side of the system, I have need
> of a spacial filtering, target discrimination and tracking subsystem,
> in order to get a clean enough signal from a particular dolphin to
> make it through a recognition process.
> 
> My thought for this is to use a head mounted line array of small
> hydrophones and a beam former. But I also wish to seperate the
> signals from multiple targets. So I'm thinking that I not only need
> several beam formers (one per target), but I also divide up the 
> spectrum into many bands and for any given target, reject bands
> which do not have a strong signal coming from the same direction
> as a particular target. Obviously, I have to aim the beam former based
> upon the current target vector as either the dolphin or the swimmer
> moves around.
> 
> So my question is this: do I put the filter bank before or after the
> beam former for a particular target ?
> 
> If I filter before, doesn't that mean I need one bank per line array
> element per target ?
> 
> What DSP chips might be able to handle this problem without needing
> Hoover Dam to power them ?
> 
> I'm not only an newbie amateur in DSP matters.
> 
> Any advice would be appreciated.
> 
> Regards,
> 
> Michael A. Hobson
> warrior (underline) mike (Y2K+1) (at) yahoo (dot) com
> 
> 
> Michael A. Hobson
> email: thesneakster (at) gmail (dot) com
> icq: #2186709
> yahoo: warrior_mike2001
Didn't you post something like this a few years back?

Get a copy of Richard O. Nielsen, "Sonar Signal Processing".  It will 
answer a bunch of questions.  You need to tell us what the frequencies 
you intend to commune with the Dolphins are.  It's not clear if you wish 
to transmit as well as recieve.  I think that power is likely to be a 
problem.  I also tend to think that an array might be too heavy.

______________________________
New DSP Code Snippets Section now Live.   Learn more about the reward program for contributors here.

Michael A. Hobson wrote:
> Hi all,
> 
> First of all, let me state that this is not a joke and I'm perfectly 
> serious.
> 
> I'm researching design options for a proposed device to facilitate
> two-way language-based communication with bottlenose dolphins. One
> of the requirements is that this device is wearable by a swimmer in 
> the water, and totally stand alone.
> 
> As I see it, for the dolphin-to-human side of the system, I have need
> of a spacial filtering, target discrimination and tracking subsystem,
> in order to get a clean enough signal from a particular dolphin to
> make it through a recognition process.
> 
> My thought for this is to use a head mounted line array of small
> hydrophones and a beam former. But I also wish to seperate the
> signals from multiple targets. So I'm thinking that I not only need
> several beam formers (one per target), but I also divide up the 
> spectrum into many bands and for any given target, reject bands
> which do not have a strong signal coming from the same direction
> as a particular target. Obviously, I have to aim the beam former based
> upon the current target vector as either the dolphin or the swimmer
> moves around.
> 
> So my question is this: do I put the filter bank before or after the
> beam former for a particular target ?
> 
> If I filter before, doesn't that mean I need one bank per line array
> element per target ?
> 
> What DSP chips might be able to handle this problem without needing
> Hoover Dam to power them ?
> 
> I'm not only an newbie amateur in DSP matters.
> 
> Any advice would be appreciated.
> 
> Regards,
> 
> Michael A. Hobson
> warrior (underline) mike (Y2K+1) (at) yahoo (dot) com
> 
> 
> Michael A. Hobson
> email: thesneakster (at) gmail (dot) com
> icq: #2186709
> yahoo: warrior_mike2001

I'd look at placing your hydrophones in a more stable location than on 
the head.  I have no strict evidence to suggest this, but I feel that 
the constant movement of the head might cause some issues with the 
beamformer.
Instead I'd look at placing an array of hydrophones on the front and 
back of the swimmer, and then use a set of beamforming 
filters/processors to combine the signals to perform the same effect.
There's quite a bit of information available for a similar situation 
using rf antenna, so there is a base of theory out there on it.
The more targets you would wish to 'track' with the system would 
determine how many beamforming processors would be required.

As for the filter bank you mention, I'm not 100% sure on what this 
filtering is for.  If it is to simply attenuate out of band signals then 
it could be placed either before the beamforming processor or after, 
though after would probably make more sense if the filter has high 
performance requirements.  Otherwise if it's a simple filter (such as 
low pass / bandpass and only order 2->4) then you could place it before 
the beamforming processor and allow it to act as an anti-aliasing filter 
for the likely digital sampled beamforming processor.

______________________________
New DSP Code Snippets Section now Live.   Learn more about the reward program for contributors here.

"Michael A. Hobson" <m...@crusader-services.com> wrote in message
news:p...@4ax.com...
> Hi all,
>
> First of all, let me state that this is not a joke and I'm perfectly
> serious.
>
> I'm researching design options for a proposed device to facilitate
> two-way language-based communication with bottlenose dolphins.

You need to buy Dragon Dictate (Dolphin edition). I find it works well
though the Dolphins have to train their voice by reading a passage of Jules
Verne. Try it and see.

McC

______________________________
New DSP Code Snippets Section now Live.   Learn more about the reward program for contributors here.