How to use the window functions on a signal in MATLAB?

[diredragon]

Homework Statement

I am suppose to write a program that compares the FFT (Fast Fourier Transform Diagrams) of a sampled signal without the use of a window function and with it. The window function should be as long as the signal and the signal should have N points, N chosen as to not cause leakage of the specter of the signal.
**The window function should be such that it's components diminish from 0 to it's end.

Homework Equations

3. The Attempt at a Solution [/B]
I am having troubles applying the window functions because i get some weird output signal from a pretty simple line of code so I'm sure I'm not getting how this is suppose to work in MATLAB. Let's say I am trying to use the triangle window function because I am not sure what is meant by the ** condition in the assignment. Maybe that the triangle function should be symmetric on the y axis?

f1 = 1000;
f2 = 1270;
fs = 15000;
F1 = 100/1500; %minimum N so i wouldn't have signal leaks on the frequency diagrams.
F2 = 127/1500;
N = 1500;
n = 0:N-1;
x = 8*cos(2*pi*F1*n)+5*cos(2*pi*F2*n);

% i have found that the triangle window function is called with triang(L)
window = triang(N); %length should be N as the signal is of length N
xwindowed=x.*window;

figure 
stem(n, x);
pause

figure 
stem(n, xwindowed);
pause

The original signal outputs like this:

and the windowed like this:

which makes no sense and is not how this is suppose to work. What don't i understand here?

 

[DrClaude]

I'm not sure what the second plot looks like it does. What are you getting for the FFTs?

 

[diredragon]

I'm not sure what the second plot looks like it does. What are you getting for the FFTs?

It turns out that the built in function returns Nx1 (for some reason) matrix and the signal is 1xN. Reshaping the result of the window function makes it work.