CMA equalizer

[fangming he]

Some parts of this message have been removed. Learn more about Nabble's security policy.

Hi, All,

For the gr_cma_equalizer_cc, GNU Radio just offer class for it. It does not provide the example  how to calculate the modulus and mu. In my opinion, the calculation of modulus should be a nonlinear step in Equalizer. It also covers high order statistics information.  

Does anyone can provide the example code for it?

Thanks a lot!

Fangming

---

Windows Live: Make it easier for your friends to see what you’re up to on Facebook.
_______________________________________________
Discuss-gnuradio mailing list
Discuss-gnuradio@...
http://lists.gnu.org/mailman/listinfo/discuss-gnuradio

[Sebastiaan Heunis]

Hi

For the CMA equaliser, the modulus refers to the desired I^2+Q^2
(modulus) value that each sample pair should have.  This is a constant
value, since any variations are due to multipath.  The algorithm tries
to remove amplitude variations given what the modulus of the signal
should be in the absence of multipath.

mu is the step size.  I don't exactly know how to calculate the
boundaries for mu to ensure convergence, but you're right, it is
related to the higher-order statistics of the signal.  I know that mu
should be very small (something like 1*10^-9) to ensure convergence.
All the times I've used it, I've used trial and error to get a value
for mu.

I implemented a CMA equaliser in Matlab that I can send you if you're
interested.

Sebastiaan

--
Sebastiaan Heunis
Radar Remote Sensing Group, University of Cape Town, South Africa
Tel:  [phone]


_______________________________________________
Discuss-gnuradio mailing list
Discuss-gnuradio@...
http://lists.gnu.org/mailman/listinfo/discuss-gnuradio