How to remove zeroes not defined by indices

Hello All,

I have a large signal data file (about 3000000x1 double) and I first ran these values under a defined signal range and replaced the non conforming values to zero.

However, some of the values of the original data also contain zeroes for which I have the corresponding indices.

1. How do I remove the other zeroes from the data but keep the zeroes corresponding to the indices?

Eg: 1. A = [ 1 2 0 0 3 0 4 0 5 6 0 9] -> Data

B = [3,6,8] -> Indices corresponding to the zeroes in A

Expected Output: [1 2 0 3 0 4 0 5 6 9]

Below is a code snippet for this logic that I tried but without success.

%Thr = Measured Data Signal (Appox. 3000000 x 1 double)
%idx = Indices for zeros to be retained in data signal

Thr2 = zeros(size(Thr));
Thr3 = Thr;
for k = 1:numel(idx)
    Thr2(idx(k)) = idx(k);
end

for k1 = 1:numel(Thr)
    if((Thr(k1)== 0)&&(k ~= Thr2(k)))
        Thr3(k1) = [];
    else
        Thr3(k1) = Thr3(k1);
    end
end

Any help would be appreciated.

Cheers!

[madhan ravi]

By the way your expect output should be

    [1
     2
     0
     3
     4
     0
     5
     0
     6
     9]

[Suraj]

Hello Madhan,

In the example, B corresponds to the indices of the values that are equal to zero and that which need to be retained.

Therefore, only the zeroes that are present in the index values 3,6,8 need to be retained and the other zeroes (In indices 4, 11] need to be removed.

Thanks!

Answer 1

[madhan ravi]

 Wanted = zeros(numel(nonzeros(A)) + numel(B), 1);
 Wanted(setdiff(1 : numel(Wanted), B)) = nonzeros(A)

[Suraj]

In the original data, there are a large number of values containing zero that are not important for the calculation and only those zeros in the original data that are present corresponding to the indices are to be kept and all other zeroes removed.

Eg: A = [0 2 0 0 0 0 0 0 0 0 0 3 5 0 0 6 2 0] - Original Data

B = [ 1 3 4] - Indices corresponding to the zeroes of original data that need to be retained

ExpOutput = [0 2 0 0 3 5 6 2]

(Zeroes that were present in the 1st, 3rd and 5th indices are retained and the rest removed)

[madhan ravi]

My two options give you the desired answer, take the time and respond.

 output = A;
 output(B) = nan;
 output = nonzeros(output);
 output(isnan(output)) = 0

[Suraj]

Perfect! Thanks for your help Madhan!