tmoshe wrote:
> Hi,
> When using the SG algorithm for equalizing, the error signal is computed
> at the end of the pll.

SG stands for old good Stochastic Gradient LMS, right?

  My problem is that the pll cannot be locked before
> the channel is reasonably equalized.

This is a difficult situation. The universal solution would be maximum 
likelihood approach, however there may be the shortcuts depending on 
what is your signal and what is your channel.


  Thus, I have a "bootstrap" where the
> equalizer need phase information(from pll) and the pll need equalized
> signal in order to work properly. 
> I still have not simulated this situation, so I need your experience
> here:
> Does such scheme suppose to work and eventually to converge,

Depending on the SNR, the initial state and the parameters, it can 
wander for a long time or it can not converge at all.


  or should I
> use other algorithm for the equalizer, (one that is not sensitive to phase
> errors)?

It does not matter how you are updating the equalizer. The adaptation of 
the equalizer is trying to compensate for ALL kinds of errors, including 
the carrier and  the symbol sync error. There is no simple way to 
distinguish the ISI from that errors.

Vladimir Vassilevsky

DSP and Mixed Signal Design Consultant

http://www.abvolt.com

Hi,
When using the SG algorithm for equalizing, the error signal is computed
at the end of the pll. My problem is that the pll cannot be locked before
the channel is reasonably equalized. Thus, I have a "bootstrap" where the
equalizer need phase information(from pll) and the pll need equalized
signal in order to work properly. 
I still have not simulated this situation, so I need your experience
here:
Does such scheme suppose to work and eventually to converge, or should I
use other algorithm for the equalizer, (one that is not sensitive to phase
errors)?
Thanks!