2
$\begingroup$

Is it possible to blindly estimate a coarse carrier frequency offset of an OFDM signal where $\lvert f_{\rm off}\rvert > \frac{\Delta f}2$, where $\Delta f$ = subcarrier spacing? Blind is defined as no a-priori knowledge of the OFDM signal parameters. However, if the useful period, Tu, is required then that can be assumed to be known.

I am wondering if there is a technique for coarse frequency estimation similar to the fine frequency estimation outlined by van de Beek et al $^1$.

$^1$J. J. van de Beek, M. Sandell, and P. O. Borjesson, “ML estimation of time and frequency offset in OFDM systems,” IEEE Trans. Signal Process., vol. 45, no. 7, pp. 1800–1805, Jul. 1997.

$\endgroup$
6
  • $\begingroup$ define "blind" here, please. If by blind you mean "without knowledge of the number of subcarriers or $\Delta f$", this might be hard. $\endgroup$ Oct 6, 2016 at 3:03
  • $\begingroup$ Yes, without any a-priori knowledge. However, if $\Delta f $ is required for coarse estimation then that is something that can be assumed to be known/measured. $\endgroup$ Oct 6, 2016 at 11:11
  • $\begingroup$ You could take advantage of the fact that OFDM signals are often symmetric about zero frequency. If the offset is greater than half a subcarrier spacing, then you might be able to discern the gross offset via simple spectral analysis (especially if you know the subcarrier spacing and can match the frequency bin size to it). Not a completely general method, but one that could work a good portion of the time. $\endgroup$
    – Jason R
    Oct 6, 2016 at 11:29
  • $\begingroup$ sadly, I didn't come up with something great just yet. Have a look at grinspector.wordpress.com/2016/07/08/week-7-ofdm-prototype and the sources quoted therein. $\endgroup$ Oct 6, 2016 at 12:59
  • $\begingroup$ @MarcusMüller I can't find the papers for the sources he is quoting $\endgroup$ Oct 6, 2016 at 13:38

0

Your Answer

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

Browse other questions tagged or ask your own question.