My receiver contains: AGC,timing recovery,carrier recovey blocks.

I am using gardner timing recovery with loop filter and farrow parabolic filter. The input to the farrow filter is the fractional delay. I am assuming the fractional delay should be constant after preamble time may be only small variations. The fractional delay does not settle in the preamble time. Between two data burst only pure carrier is sent so there any no transition for timing recovery to work. So immediately after the data burst there is lot of variations in the frac. delay. which causes timing errors. any suggestions???

  • 1
    $\begingroup$ I'm not sure this has enough information. From what I understand, you have some receiver which is causing you problems and you ask for suggestions? It is too broad. Add some more information or even some plots and diagrams of your setup and be detailed. $\endgroup$
    – Engineer
    Commented Feb 25, 2020 at 13:46

1 Answer 1


If your data isn't white or for certain TED there is a lot of jitter.

If you're using a pre-amble, why not use a least-squares or grid search for timing detection and then heavily damp your NDA detector? Or don't even bother and just lock it down until the next burst.

  • $\begingroup$ should timing recovery happen in the preamble itself ?after that timing variations should settle down. my timing recovery is taking more time than preample? could it be because of frequency offset between tx and rx as i am not taking care of the freq. offset. $\endgroup$
    – nancy
    Commented Feb 26, 2020 at 4:55
  • $\begingroup$ Yes, IIRC the transmission systems that have a preamble all include timing recovery. $\endgroup$ Commented Feb 26, 2020 at 6:38

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.