Empirically Optimized FFT bin interpolator

robert bristow-johnson wrote:
> Ron N. wrote:

   ...

>> The better assumption in this case is to assume the dft
>> included a rectangular window
> 
> or better yet, that the rectangular window was applied to the stream of
> data before it was passed to the DFT (which periodically extended it).

   ...

I think the terms we use mislead some people. It's easy to wrongly 
assume that "window" == "tapered window", so that a rectangular window 
is no window at all. The windows in my house are rectangular, but I 
can't see through the walls around them. For a DFT, a window is any 
selection of part of the data, which is to say, a rejection of some of 
the data that might have been collected. Whether the window alters the 
size of any of any of the samples is not relevant to its existence.

Jerry
-- 
Engineering is the art of making what you want from things you can get.
�����������������������������������������������������������������������