what does sin(pi*[-Fs:1/Fs:Fs] )mean?

It's just a sine, but I'm not sure why they have coded up that way. It has a period of 2 so depending on what Fs is you can tell how many samples constitute 1 period.

For example:

    Fs = 1000;
    y = sin(pi*[-Fs:1/Fs:Fs]);

It has a period of 2000 samples

plot(y(1:2000));

The fact that it looks like they are using a frequency variable for the sine makes me think they really want a sinc() function, which is not the same.

2 个评论
显示无隐藏无

Image Analyst 2012-11-27

在 MATLAB Online 中打开

Shradha, it probably would have been a little easier to understand if they'd used spaces and parentheses, like this:

-Fs : (1/Fx) : Fs

It's basically (startingValue) : (stepValue) : (endingValue). Of course an alternative is the linspace() function:

numberOfSteps = 2 * (Fs ^ 2);
vector = linspace(-Fs, Fs, numberOfSteps);

used mainly when you want to specify the number of steps instead of the step spacing. You can use whichever style suits you.

Wayne, I don't see how they might have wanted the sinc. It looks like a pretty standard sine to me, just like in Wikipedia: http://en.wikipedia.org/wiki/Sine_wave sine(omega*t) where omega is 2*pi*frequency.

Wayne King 2012-11-27

The fact that they are using a notation usually reserved for a frequency variable, Fs, made me question whether they were trying to approximate the Fourier transform a rectangle in time (a sinc in the Fourier domain). Even though they are using a conventional DT, 1/FS, to have a time vector run from -Fs to Fs looked a bit unconventional to me.

请先登录，再进行评论。