Beamforming with 4x4 Microphone Array

Hello,

I have assembled a 4x4 Microphone Array to collect heart sounds. This device collects 16 analog signals through an MCC DAQ. I would like to perform some sort of beamforming on these signals to steer the microphone towards the desired sounds.

What is the most efficient way to perform beamforming on my signals using the "Frost Beamformer" from the Phased Array Toolbox?

To simulate I am using sample heart sounds from a website and adding random noise. Assume

heart = [A A A A A A A A A A A A A A A A];

are the 16 signals collected from my hardware.

--------------------------------------------------------

Here is my code so far:

hmic=phased.OmnidirectionalMicrophoneElement;
ha=phased.URA([4 4],[0.0635 0.0635],'Element',hmic);
c=340;

rs = RandStream.create('mt19937ar','Seed',2008);
noisePwr = 1e-4; % noise power

fs=22257;

NSampPerFrame = 1000;
NTSample = 23000;
voice_dft = zeros(NTSample,1);

%DATA
A = data(1:23000,:);
heart = [A A A A A A A A A A A A A A A A];

for m = 1:NSampPerFrame:NTSample
    sig_idx = m:m+NSampPerFrame-1;
    x = A([sig_idx(1) sig_idx(end)]);
    y = x + sqrt(noisePwr)*rand(rs,NSampPerFrame,16);

      sigArray(sig_idx,:) = y;
      voice_dft(sig_idx) = x1;
  end

angSteer = [0;0];

%BEAMFORM
hsig=dsp.SignalReader('Signal',y,'SamplesPerFrame',NSampPerFrame);
bf=phased.FrostBeamformer('SensorArray',ha,'SampleRate',fs,...
   'PropagationSpeed',c,'FilterLength',100,'DirectionSource','Input port');

frostout=zeros(NTSample,16);
for m=1:NSampPerFrame:NTSample
   frostout(m:m+NSampPerFrame-1,:)=step(bf,step(hsig),angSteer);
end

%soundsc(frostout,fs,16)
wavwrite(frostout,'Frostbeamformed real data');
plot(t,frostout);
xlabel('Time (sec)'); ylabel ('Amplitude (V)');
title('Frost Beamformer Output');

Another note: I receive an error when I try to run the above code which I am also unsure about:

Error using  + 
Matrix dimensions must agree.

Error in frost (line 21)
    y = x + sqrt(noisePwr)*rand(rs,NSampPerFrame,16);

Answer 1

[Honglei Chen]

Not sure what the dimension is for your variable heart and A, but

x = A([sig_idx(1) sig_idx(end)]);

seems to suggest that x is just a length-2 vector. That's why it is does not match the noise which has 16 columns.

Answer 2

[Zeynep Ertekin]

Hi,

I need a 2d or 3d sound source localization code with command load; can anyone please help me. Any help will be highly apprecited.