Plot multiple results using for loop

I'm trying to plot raw and window 10, 50 and 100 all in one graph. I get:

Error using plot

Vectors must be the same length.

Error in moving_avg_filter (line 34)

plot(t,filtered_z)

When I ran the code with only plot(filtered_z). In the workspace it showed that filtered_z is 1x2296 so I tried to set t the same. However, when I ran it using plot(t,filtered_z) it didn't work and the values in the workspace changed to this:

Also, I tried placing the plot outside the for loop and it worked, but it will only give the last loop. So, how can I fix this?

Below is my code.

clear
clc
%You will need to add the logic for plotting all window 
%sizes and FFT as instructed in the lab document

%Read data in
data = csvread('z_flip_data_60BPM.csv');

%Assign columns from csv
z_raw = data(:,2); %Raw signal from data collection
time = data(:,1)/1000; %time converted to seconds

plot(time,z_raw,'r');
hold on
%Change window size as required
window = [10 50 100]; 

%sanity check to avoid improper window size settings
if window>length(z_raw)
	return;
end


%Indexing for the filtered array
j = 1;
t = time(1:2296,1);

%Moving average window filter + basic plot
for i = window+1:(length(z_raw))
	filtered_z(j) = mean(z_raw(i-window:i));
	j = j+1;
    plot(t,filtered_z)
    hold on
end

Answer 1

[Deepak]

Hi @rex a,

To fix the issue of mismatched vector lengths when plotting the filtered signals, we need to ensure that the time vector matches the length of the filtered data for each window size. This can be achieved by creating a new “filtered_time” vector that starts at the index corresponding to the end of window, ensuring it aligns with the shorter “filtered_z” vector.

Next, by iterating over each window size and plotting the results within the loop, we can maintain all plots on the same graph.

Additionally, using the “DisplayName” property in the “plot” function allows for the dynamic labelling of each line, and calling the “legend show” after plotting displays a legend that helps distinguish between raw and filtered signals.

Please find below the updated MATLAB code with the required changes:

data = csvread('z_flip_data_60BPM.csv'); 

z_raw = data(:,2); % Raw signal from data collection 
time = data(:,1)/1000; % Time converted to seconds 

figure; 
plot(time, z_raw, 'r'); 
hold on 

windows = [10, 50, 100]; 

% Loop over each window size 
for w = windows 
    % Initialize filtered_z for the current window size 
    filtered_z = zeros(1, length(z_raw) - w); 
    
    % Moving average window filter 
    for i = w+1:length(z_raw) 
        filtered_z(i-w) = mean(z_raw(i-w:i)); 
    end 
    
    % Adjust time vector for plotting filtered data 
    filtered_time = time(w+1:end); 
    
    % Plot the filtered data 
    plot(filtered_time, filtered_z, 'DisplayName', sprintf('Window %d', w)); 
end 

legend show 
xlabel('Time (s)') 
ylabel('Amplitude') 
title('Raw and Filtered Signals') 

hold off 

Please find attached the documentation of functions used for reference:

mean - www.mathworks.com/help/matlab/ref/double.mean.html

plot - www.mathworks.com/help/matlab/ref/plot.html

legend - www.mathworks.com/help/matlab/ref/legend.html

I hope this assists in resolving the issue.