This does the same thing,, except that instead of returning a cell array of the individual points, it takes the mean of each interval and returns it —
clear;
Uzp = unzip('Data 2.zip');
data = readmatrix('Data 2');
t = seconds(data(:,1));
ixv = ceil(data(:,1) + 1E-12);
secv = accumarray(ixv, (1:size(data,1)).', [], @(x){mean(data(x,[2 3]))});
secm = cell2mat(secv); % Create Matrix From Cell Array
% A2 = reshape(cell2mat(secv).', 2,10,[]); % Not Needed Here
tv = unique(ixv);
figure
plot(tv, secm, '-')
ylabel('Phase (°)')
grid
xlabel('Time')
xlim([tv(1) tv(end)])
xlim([0 86400])
figure
plot(tv, secm, '-')
ylabel('Phase (°)')
grid
xlabel('Time')
xlim([tv(1) tv(end)])
xlim([0 1E+2])
figure
stairs(tv, secm)
ylabel('Phase (°)')
grid
xlabel('Time')
xlim([tv(1) tv(end)])
xlim([0 1E+2])
The difference here is that movmean creates a moving average of adjacent elements. This code creates the mean of consecutive 10-element segments of the signal, so not a moving average and instead so sort of average of adjacent 10-element blocks of the data.
NOTE — The ‘secm’ matrix has a row length that is 1/10 the length of ‘data’. So it does what you want it to do.
EDIT — Re-ran code.
.
