finding the exact points of Zero crossing using interpolation method

Hi guys,

Im implementing in matlab zero crossing points to my signal , frequency sampling is 2048Khz.

my matlab code that Im using is:

y= % * my signal input*  it's like sinusoidal signal
x=1:length(y);
fs=2048e3;                                                                                                       % Sampling Frequency
n=length(y);                                                                                                     % Length of my signal
t = (0:(n-1))*(1/fs);                                                                                            % Time appointed for every sample, first sample at time t=0 , next tn=t0+n*(1/fs)
x=t;
zci = @(v) find(v(:).*circshift(v(:), [-1 0]) <= 0);
% Returns Approximate Zero-Crossing Indices Of Argument Vector
dy = zci(y);
% Indices of Approximate Zero-Crossings
for k1 = 1:size(dy,1)-1
    b = [[1;1] [x(dy(k1)); x(dy(k1)+1)]]\[y(dy(k1)); y(dy(k1)+1)];
    % Linear Fit Near Zero-Crossings
    x0(k1) = -b(1)/b(2);
    % Interpolate ‘Exact’ Zero Crossing
    mb(:,k1) = b;
    % Store Parameter Estimates (Optional)
end
figure(1)
plot(x, y, '-r')
hold on
plot(x(dy), y(dy), 'bp')
hold off
grid

the output I get is this photo below but the zero crossing point isn't correct - the points of zero crossing isn't accurate, isn't the exact zero crossing points !! .. the points of zero crossing should their positions be on the line of zero - implicitly the zero crossing points should be the intersections points of x axis-:

Could anyone please help me how can I implement in matlab zero crossing points of exact zero crossing points?! thanks alot

[Jimmy cho]

I've already seen posts and thread of @Star Strider answers but Im still not solving my problem.

Appreciated for any help guys!

Answer 1

[Star Strider]

You are plotting the values of the data at the indices that ‘zci’ returns, not the interpolated aero-crossings.

It would be best to plot the linearly-interpolated x-coordinate values ‘x0’:

plot(x0, zeros(size(x0)), 'o')

Use whatever marker you want.

That should work.

[Jimmy cho]

Understand ! , thanks alot for your answer, appreciated !!

[Star Strider]

As always, my pleasure!

(I likely wrote that original code!)

[Jimmy cho]

could you please help me over here? Im stuck about two month on this !!

https://www.mathworks.com/matlabcentral/answers/619178-likelihood-decision-in-viterbi-decoding-matlab?s_tid=prof_contriblnk

appreciated

[Star Strider]

My knowledge of digital communications engineering is at this point infinitesimal. (My last exposure to it was decades ago, and not much then.) I would like to help, however that is far outside my areas of expertise. (My apologies.)

It is highly likely that someone with necessary expertise will see it and respond, although that could take hours to a few days. In the interim, search the File Exchange to see what may be available.