You may want to try something like below
%% Signal Model
t = linspace(0,1.5,1000);
mu = [1 5 9 13]/10; % position
amp = [3 2 3 1]; % amplitude
sigma = [2 3 6 4]/100; % width
x = 0;
for n = 1:length(mu)
x = x + amp(n)*exp(-((t - mu(n))/sigma(n)).^2);
end
figure, plot(t,x), xlabel('Time [sec]'), ylabel('Amp.'), grid("on")
%% Find Peaks
[pks,locs,w] = findpeaks(x,t,...
'Annotate','extents',...
'WidthReference','halfheight');
%% Plot Peak Area
figure
plot(t,x,'LineWidth',2), xlabel('Time [sec]'), ylabel('Amp.'), grid("on")
ylim([min(x)-0.5,max(x)+0.5])
hold("on")
for n = 1:length(pks)
tstart = locs(n)-w(n);
tend = locs(n)+w(n);
ind = sort(find((t >= tstart) & (t <= tend)));
pax = [t(ind(1)),t(ind(1)),t(ind(end)),t(ind(end))];
pay = [min(x([ind(1),ind(end)])) pks(n) pks(n) min(x([ind(1),ind(end)]))];
patch(pax,pay,'yellow','FaceAlpha',0.3)
end
We can play with the factors of the width to cover the desired area manually and approximate the Riemann integral using functions like sum or trapz for each of the highlighted area.
