Find the local maxima and minima of a data set by using the functions diff, sign, find

I have a vector containing the time of analysis and three other vectors containing the cellule concentrations. This system shows an oscillatory behavior. I need to plot the concentrations in function of the time -> plot(t,C) and highlighting on the graph the local maxima and minima with a circle -> plot(t,Min,'o'). I am asked to use the functions diff, sign, find (logical index). I am able to solve this with findpeaks or a for loops. I know it is possible to determine the maxima and minima with sign(diff(C)) but then i don't know how to do.

Any suggestion?

Thank you very much, Michael Daldini

Answer 1

[Star Strider]

This should get you started:

t = linspace(0, 10, 1000);                      % Time
C = 2 + sin(2*pi*t + cos(5*pi+t)*pi);           % Signal

dC = diff([0, C])/(t(2)-t(1));                  % Approximate Derivative
dCzx = dC .* circshift(dC, [0 -1]);             % Negative Values => Zero Crossings (Mostly)
zx_idx = find(dCzx <= 0);                       % Derivative Zeros Are Maxima, Minima Of ‘C’

figure(1)
plot(t, C)
hold on
plot(t(zx_idx), C(zx_idx), 'or')
hold off
grid
axis([xlim    0  5])

Note that the dCzx calculation results in a zero-crossing at the end that really is not one. I will let you decide how best to deal with that if it occurs with your data, since that’s part of the fun or real-world data analysis.