Allen wrote:
> Dear all,
>
> I read the thread started by rider
> two questions
> 1. what will happen to Gardner TED if I give it 10 sample per symbol
> instead 2 sample per symbol ?
>
>
> 2. if i want to down sample from 10 => 2
> where should i put the decimator
> i suggest to put in front of interpolator instead of TED
> what do you think
>
>
> my TRL system
> input => decimator => interpolator => TED => NCO => interpolator
>
>
> thank you for your time
>
There is no need to decimate. If you have 10 samples per symbol, just
use every 5th sample for the Gardner algorithm.

Regards,
Steve