Filter noise in logical vector

I have a logical vector that is meant to run over time (ms). It looks like this:

0000000000000110000000000001111111111111111111111111111111111111100000000000000000001000000000
'Time in ms--->'

In this example I only want a signal that ran 'true' longer then 3ms. The rest I want filtered back to zero's becuase I still need to plot it against time.

How should I do this?

Answer 1

[KALYAN ACHARJYA]

Do Modification: Just Hint here (May Be)

loop
data_generation(i)=present_data;
if present_data==1
tic
while toc==
end 
end

Answer 2

[Agustin Canalis]

An interesting choice is to use movmin and movmax, which work along a sliding window in your vector:

A = [0 0 0 0 1 1 0 0 0 0 0 1 1 1 1 1 1 1 1 1 0 0 1 1 1 1 1 1 1 0 0 1 0 1 0 0]

window_size = 3 % or scale to your needs

B = movmin(A,window_size)

C = movmax(B,window_size)

imshow([A;B;C]) % This is just to graph the result

Here's the result:

The idea is that movmin "triggers" when the sliding window contains 3 consecutive 1's, otherwise the minimum is 0.

However, this also clips useful portions of your signal at the start and end, which is why I have to use movmax to add them back.

Because it's vectorized I guess the performance is better than in a loop, but I haven't tested it.

Cheers!