Well the problem statement has your answer. From the definition of the fibonacci number, you have: F(1) = 1, F(2) = 1, F(N) = F(N-1) + F(N-2).
Therefore, you need two variables to track the two previous numbers. You need a variable to track the current number that has been calculated, and should that reach your 'm', the for loop ends.
(I suppose you need to do it with a for loop as an assignment, as a while loop would be cleaner and easier to write for this problem)
And at every step of the loop after F(k) has been calculated it's simple to proceed as F(K+1) = F(K) + F(K-1), which means that the variable that held F(K-2) needs to contain what the variable that held F(K-1) has.
Or you could use a recursive function which maps the problem and the solution of calculating a fibonacci number more closely than any non-recursive algorithm does.
