🚀 go-pugleaf

I've been thinking about using SSB techniques to shift the spectrum
of audio signals in the audio range by frequencies ranging from 0.2
to 20 Hz.  The obvious way to do this with analog electronics is with
the phasing method:

           ____________
          |            |
          |            |------o A
   In o---|  Phase     |
          |  Shift     |            Phase difference of 90 degrees
          |            |------o B
          |____________|

           ____________
          |            |
          |            |------o X
          |   Sin/Cos  |
          | Oscillator |            Phase difference of 90 degrees
          |            |------o Y
          |____________|


Output is A*X + B*Y or A*X - B*Y, depending on
whether you want the upper or lower sideband.

The design of phase shift networks usually involves collections
of all-pass filters whose differential phase shift is 90 degrees
over a wide range of frequency.  I understand that this approximates
a Hilbert transform.

My questions are:

   How are the corner frequencies of the all-pass networks derived?
   I would like to design an analog phase shift network that works
   well over the entire audio range, say 20-20KHz.

   How can a similar thing be done with a digital signal processor?
   Assuming a DSP is available, what's the best way to accomplish
   this task?
--
Robert Kelley, Software Engineer                      503-626-1278
Mentor Graphics Corp., 8500 SW Creekside Place, Beaverton OR 97005
...!{sequent,tessi,apollo}!mntgfx!bobk   OR    bobk@pdx.MENTOR.COM
These are my opinions, & not necessarily those of Mentor Graphics.