xy - it seems as if your code is iterating over the y array and looking for those elements which are a maximum with respect to its neighbours
if (y(i-1)<y(i) && y(i+1)<y(i))
When this is true, what do you want to save: that maximum value or its index into y? If the former, then you should be saving y(i) and if the latter, then you should be saving i. Something like
maximumValues = [maximumValues y(i)]; % (1)
or
maximumIdcs = [maximumIdcs ; i] % (2)
where we have initialized either array to the empty matrix (like you have done for maxim). Either of the above would replace your code
for j=1:(j+1)
maxim(j) = y(i)
end
which I don't quite follow. The above iterates from 1 to j+1, replacing the first j+1 elements of maxim with the same y(i). This is why your maxim array has identical values...because they are getting reset (or initialized) to same y(i). I think that you just need to replace this with (1) or (2) from above and you will be have what you are looking for.
As an aside, it is better to not use i and j as indexing variables as MATLAB uses both of these to represent the imaginary number.
