How to filter and remove unwanted noise from muscle contraction reading data in MATLAB

I have a project where I need to remove unwanted noise in MATLAB data. It is a lot of data and contains a lot of noise. How can I make this data readable? I am very new to MATLAB and not fimiliar with most of the commands.

Thank you!

[Mathieu NOE]

hello

can you share some data file ?

tx

[TheNewGuy]

Yes. There is a little bit more than just removing noise that I need. I was given two sets of data collected from an electrode and an oscilloscope. There are electrical measurements of the right and left bicep for each. The two sets of electrical data are different trials but taken over a different time interval. I am trying to determine which bicep is stronger based on the two sets of measurements. It is unknown which Bicep is stronger. Sample rate is 8000Hz. The first set is over 1 second at 8000Hz and the second is over 10 seconds at 8000Hz. Thank you!

Answer 1

[Alagu Sankar Esakkiappan]

Hello!

I see that you're trying to remove unwanted noise from your dataset in RightvsLeftBicep.xlsx . There are many ways to filter such data using MATLAB. One such method is smoothdata which uses weighted averaging among others to reduce noise. There are many filtering methods in smoothdata and different window periods you may try out to see which one fits best for your use case. You may refer to the following code snippet for reference:

warning('OFF', 'MATLAB:table:ModifiedAndSavedVarnames'); % To Suppress Warning due to incompatible Column Names in Dataset
exerciseTable = readtable( "RightvsLeftBicep.xlsx"); % You may try avoiding Spaces in 1st row of DataSet as good practice
t = exerciseTable{:,1};
rightBicep = exerciseTable{:,2};
leftBicep = exerciseTable{:,3};
% Every single entry as output of smoothdata is a result of 1000 Sample points.
% You may try reducing or increasing them as per your use case to further
% reduce sharp peaks and troughs.
plot(t,smoothdata(rightBicep,'sgolay',1000)); 
hold on;
% Alternatively, You may also explore other
% methods in smoothdata documentation
plot(t,smoothdata(leftBicep,'sgolay',1000));
legend("Right Bicep","Left Bicep");

[TheNewGuy]

Thank you! This is what I needed.