How to add baseline drift to an ECG signal

I am at the beginning of the study of ECG signal processing and I am at the start of learning MATLAB. I cannot tell if I am doing the correct operations.

the image presented above presents the operations to be performed to add the baseline drift to the starting signal.

I don't know if it is wrong or not. But when I plot the signal with the addition of the baseline drift, it seems almost unchanged.

To perform these operations I used the signal found in this database (name of the file: 100m.mat)

link: https://physionet.org/content/mitdb/1.0.0/

(to download the file in the .mat version follow the link: https://archive.physionet.org/cgi-bin/atm/ATM to download the .mat version file follow the link: https://archive.physionet.org/cgi-bin/atm/ATM and select the database called "MIT-BIH Arrhythmia Database (mitdb) and select 100 as Record, than Toolbox and selcet export as .mat")

The one below is the code I wrote:

%% Original ECG Signal
load('100m.mat');
original_signal = val(1,:);
fs = 360; % Hz
% T = 1/fs; 
L = length(original_signal);
t = (0 : L - 1)/fs;

plot(t, original_signal);
title('plot of the original ECG signal');
xlabel ('time [sec]');
ylabel ('ECG Amplitute [mV]');
grid on

%% Add Noise to the ECG signal
Noise = awgn(original_signal, 20, 'measured');
plot(t, Noise, 'b');
grid on
xlabel('time [sec]');
ylabel('ECG Amplitute [mV]');
title('ECG signal with added noise');

%% Add Baseline drift to ECG signal
x = linspace(0,2*pi,L);
A = 0.8; 
N = 60; 
BaselineDrift = A*cos(x./N);
subplot(2,1,1)
plot(t, Noise); % Noise is the original signal with noise add to it
xlabel('time [sec]');
ylabel('ECG Amplitute [mV]');
title("Noisy ECG Signal");

subplot(2,1,2)
Dirty_Signal = Noise + BaselineDrift;
plot(t, Dirty_Signal)
xlabel('time [sec]');
ylabel('ECG Amplitute [mV]');
title("Corrupted ECG Signal (Noise + Baseline Drift)");

Thanks in advance for the help

Answer 1

[Image Analyst]

That's not all the code. Where do you define L, t, and Noise?

Try

minDriftOffset = 0; % Whatever
maxDriftOffset = 10; % Whatever.
BaselineDrift = linspace(minDriftOffset, maxDriftOffset, length(Noise));
Dirty_Signal = Noise + BaselineDrift;

[Sabaudian]

I am sorry I am going to add the rest of the code and I'll try to implement your suggestion

[Image Analyst]

And attach '100m.mat' with the paperclip icon.

[Sabaudian]

I have attach the file

[Image Analyst]

Sorry I couldn't run it. awgn() requires the Communications Toolbox. But did you try my suggestion? If not, why not?

My code had the drift be a linear ramp while

BaselineDrift = A*cos(x./N);

has the drift be a cosine shape. You could combine both with

minDriftOffset = 0; % Whatever
maxDriftOffset = 10; % Whatever.
BaselineDrift = A * cos(x ./ length(Noise)) + linspace(minDriftOffset, maxDriftOffset, length(Noise));
Dirty_Signal = Noise + BaselineDrift;

[Sabaudian]

I try it and it works thank. I hope it is consistent with what is presented in the document I plot above.