Forums

equalization of 'non-continuous' sample

Started by cl7teckie March 11, 2010
I am trying to build a 'non-continuous' 12-tap FFE (LMS Blind) equalizer in
software, hoping to remove to some extent group delay in a satellite link.
I am doing in software, as I am ONLY keen to know the SNR after
equalization. The reading are in chunks of 512 x I/Q samples. 

Technically, does this even work, due to the fact that it is
non-continuous. I do not need a perfect system, just some improvement will
go a long way.



cl7teckie wrote:

> I am trying to build a 'non-continuous' 12-tap FFE (LMS Blind) equalizer in > software, hoping to remove to some extent group delay in a satellite link. > I am doing in software, as I am ONLY keen to know the SNR after > equalization. The reading are in chunks of 512 x I/Q samples. > > Technically, does this even work, due to the fact that it is > non-continuous. I do not need a perfect system, just some improvement will > go a long way.
That could work. Why not? How well it would work - it depends... Vladimir Vassilevsky DSP and Mixed Signal Design Consultant http://www.abvolt.com
cl7teckie wrote:
> I am trying to build a 'non-continuous' 12-tap FFE (LMS Blind) equalizer in > software, hoping to remove to some extent group delay in a satellite link. > I am doing in software, as I am ONLY keen to know the SNR after > equalization. The reading are in chunks of 512 x I/Q samples. > > Technically, does this even work, due to the fact that it is > non-continuous. I do not need a perfect system, just some improvement will > go a long way.
I hope you mean "equalize", not "remove". Jerry -- Why am I in a handbasket? Where are we going? �����������������������������������������������������������������������