1 Mar
2010
1 Mar
'10
6:56 a.m.
On Sun, Feb 28, 2010 at 08:35:37PM -0500, Gene Heskett wrote:
On Sunday 28 February 2010, Gordon JC Pearce wrote:
So did mine Gordon, heck, while I know what it is, and that its symetrical, I
could write an FFT or even a butterfly transform with a gun to my head. But
I have sure made heavy use of gear that can do it over the years.
The discrete form is a set of two FIRs I[i] and Q[i]
with
Q[i] = 0 for even i
Q[i] = 2 / (pi * i) for odd i
for i = -N...N, delayed by N samples to make it causal,
and I[i] is just a delay of N samples.
Q is an finite-lenght approximation of the real thing,
so its FR is not entirely flat. If it matters you can
modify I for the same FR to compensate.
Ciao,
--
FA
O tu, che porte, correndo si ?
E guerra e morte !