Compute RMS value over 1-second intervals

Trying to get RMS 1s value from a excell file which contains 2 collums of data . collum 1 time and collum 2 is accelaration. I appled Wk filter and code runs fine but now i wanted to Compute RMS value over 1-second intervals but its howing error.

% Compute RMS value over 1-second intervals
windowSize = round(fs); % Number of samples in one second
a_rms = movrms(ac, windowSize); (ERROR in this line)

Answer 1

[Mathieu NOE]

hello

I am not aware of a matlab function called movrms (but there is a movmean and movsum)

here a code that does the same thing with the possibilty also to add overlap

Edit : improved code

% dummy data
n = 1000;
Fs = 100;
dt = 1/Fs;
t=(0:n-1)*dt;
y = max(0.707,abs(cos(t))+0.1*rand(size(t)));

buffer = Fs;  % 1 second buffer 
overlap = round(0.5*Fs); % overlap (here 50 % of buffer size)

[t_rms,y_rms] = my_rms(t,y,buffer,overlap);

figure(1),
plot(t,y,t_rms,y_rms,'r-*');
legend('raw data','1 s rms');

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
function [t_rms,x_rms] = my_rms(t,x,buffer,overlap)

%%%% main loop %%%%
m = length(x);
dt = mean(diff(t));
shift = buffer-overlap;    % nb of samples between 2 contiguous buffers  

for ci=1:fix((m-buffer)/shift +1)
    start_index = 1+(ci-1)*shift;
    stop_index = min(start_index+ buffer-1,m);
    t_rms(ci) = dt*(start_index+stop_index)/2;  % time (centered in buffer)
    x_rms(ci) = sqrt(mean(x(start_index:stop_index).^2));
end
   
    
end

[Mathieu NOE]

Hello

Problem solved ?

Answer 2

[Star Strider]

Try this —

movrms = @(A,k) sqrt(movmean(A.^2,k));

Fs = 1000;
T = 10;
t = linspace(0, T*Fs, T*Fs+1).';
s = randn(size(t));

srms = movrms(s,round(Fs));

figure
plot(t,s, 'DisplayName','Signal')
hold on
plot(t, srms, '-r', 'DisplayName','RMS')
hold off
grid
legend('Location','best')

s = sin(2*pi*t/10);

srms = movrms(s,round(Fs));
Check = srms([1 end-1000])

Check = 2×1

0.7071 0.7071

figure
plot(t,s, 'DisplayName','Signal')
hold on
plot(t, srms, '-r', 'DisplayName','RMS')
hold off
grid
legend('Location','best')
xlim([0 200])

See the documentation on movmean for details. My ‘movrms’ function can be changed to accommodate other arguments to movmean by changing the ‘movrms’ argument list accordingly.

This gives the correct result, the RMS value of a sine being .

.