woo hoo! physics lecture time! since everyone else answered the practical side, I'll try to explain what's going on behind the scene.
-note: if you don't care about sampling and recovery, and other aspects of applied digital electronics, you can skip this post and save yourself a headache.
Aliasing is an undesirable by-product of all digital to analog conversions. Because all data is stored at discrete values, you get that "jaggedness" in what would otherwise be a continuous function.
For example, imagine a sine wave. When sampled and restored, you would have something that looks like a bunch of ascending and descending stairs. From a mathematical standpoint, this is a construct of an infinite number of odd harmonics (multiples of the fundamental frequenct of our sine wave).
An anti-aliasing filter is a low-pass filter that (in theory) begins to operate just above the frequency of our original sine wave. This removes all those extra, unwanted frequencies and gives us a very close approximation of our original input signal.
I apologize if this isn't the most thorough explantion. I'm trying to pack about 4 weeks of digital theory lectures into a short little paragraph. Let me know if you need clarification.
[edit]I know all that stuff, but I still can't spell.[/edit]
__________________
mmmm.... pudding
Last edited by digby; 07-22-2003 at 05:59 AM..
|