Electronic Music mailing list archives

James Chandler Jr wrote:
> There was a 'pretty good' IIR approximation Hilbert in CSound source 
> last time I looked (a few years ago). Maybe still in there.
>
> The CSound Hilbert was a digital emulation of an analog Hilbert which 
> was published by Bernie Hutchins in his Electronotes magazine long ago.
>
> It sends the signal thru two legs of cascaded first-order phase 
> shifters. The 'staggered' FC's of the two phase shifter chains, 
> resulted in the output of one phase shifter cascade, a pretty good 
> approximation of 90 degrees shifted against the other phase shifter 
> cascade output.
>
> But neither output had a Hilbert relationship to the original signal.
>
> That is how it accomplished the requisite delay so it could run in 
> realtime. One phase shift leg was just delayed more than the other 
> phase shift leg.
>
> Pretty ingenious, I thought.
>
This sounds like the filters I got from Olli Niemitalo (yes, google 
confirmed that)

http://www.student.oulu.fi/~oniemita/dsp/hilbert/

The explanation is pretty good and interesting. I used them for a 
"complex ring modulator" with just one sideband, and also fiddled around 
for envelope detection (in some cases, adding the absolute values 90 deg 
shifted signals reduces the effect of envelope detector riding the 
waveform).

hth,
Vesa

--
dupswapdrop -- the music-dsp mailing list and website: 
subscription info, FAQ, source code archive, list archive, book reviews, dsp links 
http://music.columbia.edu/cmc/music-dsp 
http://music.columbia.edu/mailman/listinfo/music-dsp