Richard_K wrote:

> As I understood, the eigenvalue spread of the input signal will affect the
> performance of LMS MMSE algorithms. But will the eigenvalue spread affect
> the performance of LMS MBER and AMBER algorithms?

Of course it will. The problem is fundamental: poles are compensated by 
zeroes and zeroes are compensated by poles, but not vice versa. It does 
not matter what is your convergence criteria.


Vladimir Vassilevsky

DSP and Mixed Signal Design Consultant

http://www.abvolt.com

As I understood, the eigenvalue spread of the input signal will affect the
performance of LMS MMSE algorithms. But will the eigenvalue spread affect
the performance of LMS MBER and AMBER algorithms?

Thanks.