discuss-gnuradio
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Discuss-gnuradio] Fwd: ofdm demodulation


From: Karan Talasila
Subject: [Discuss-gnuradio] Fwd: ofdm demodulation
Date: Thu, 18 Apr 2013 14:49:03 +0530



---------- Forwarded message ----------
From: Karan Talasila <address@hidden>
Date: Thu, Apr 18, 2013 at 2:48 PM
Subject: Re: [Discuss-gnuradio] ofdm demodulation
To: Brian Padalino <address@hidden>


I understand the intuitive way in which first the phase offset is found. from that you can find finefrequency offset and timing offset. But I  am looking at how all this is done in the code using appropriate equations, especially how coarse frequency offset is found using the math. The paper says there will be a factor of 2z/T that you should consider when the offset is greater than half the subcarrier spacing. i.e. integer frequency offset. Hopefully a page on how the sync is being implemented in the code would help. I have to blame my poor c++ skills for all this pain.


On Thu, Apr 18, 2013 at 1:44 AM, Brian Padalino <address@hidden> wrote:
On Wed, Apr 17, 2013 at 6:30 AM, Karan Talasila <address@hidden> wrote:
Hi,
 

I am trying to understand ofdm demodulation using schmidl and cox paper. Can anybody explain how a pn sequence on even frequencies and 0 on odd frequencies help in the estimation of timing and frequency offset.


Hi Karan,

I haven't read the paper, but here is what I know from the description of the scenario.  Someone who has more OFDM experience can probably speak with a little more authority.

If you skip every other bin in the frequency domain, the time domain signal is actually repeated during the whole OFDM symbol. Since the time domain information is repeated, you can perform a correlation with a delayed version of the symbol coming in for some rough synchronization of the preamble.  Also, the phase difference of that result will be the average frequency offset over that number of samples, giving you a decent frequency offset estimate as well.

At least, that's what I think would be useful.

Does that make sense to you?

Brian
 
--
Regards
Karan Talasila

_______________________________________________
Discuss-gnuradio mailing list
address@hidden
https://lists.gnu.org/mailman/listinfo/discuss-gnuradio





--
Regards
Karan Talasila



--
Regards
Karan Talasila

reply via email to

[Prev in Thread] Current Thread [Next in Thread]