Yes, you are right. But, why is it so? If everything else is right, I should have been getting a similar graph without averaging it as well; isn't it?
Can it happen because of non-equal data? But, I am feeding in same bytes (Random Uniform Source: Min = Max = Seed = 255), and further also, if I employ QAM, which should give me same Pre-OFDM data, then after OFDM modulation, again, I should be getting the same symbols which should lead to such peaks in the FFT output.
My problem is that I have to analyse each sub-carrier separately, where average case won't help.
Second thing which I fail to understand here are the large peaks in the relatively empty part (unoccupied part of the OFDM symbol), and in the middle.
Can you please help me with this.
On Sun, Jan 3, 2016 at 8:16 PM, Ron Economos <address@hidden> wrote:
If you turn the FFT averaging up to maximum and wait a little while, the carriers will stand out.
Ron
On 01/03/2016 05:03 PM, Aditya Virmani wrote:
Hello
Yes, I think this is the same thing I am trying to do, but I am not getting the results I expected. Can you please share the GRC file for this. In my execution, I was getting null signals.You can see my setup in the attached file. It's a rather simple version just for illustration. I don't understand why that is not working.
On Sun, Jan 3, 2016 at 7:45 PM, Ron Economos <address@hidden> wrote:
If you want to see individual carriers, you have to use a high resolution FFT. Here's a 262144 point FFT on a DVB-T2 OFDM signal with a 1K FFT (853 active carriers). The carriers are spaced about 9 kHz. Stronger tones are pilots.
https://twitter.com/drmpeg/status/670769195160858624
Ron
On 01/03/2016 04:34 PM, Aditya Virmani wrote:
Hello sir
Thank you for your reply. Well, I am trying to do some additional operation to the receiver side of the OFDM transceiver (rx_OFDM).I am intending to detect a doppler shift in the received signal, for which I am trying to do an FFT over symbols collected in 1 second.To test the case, I tried to do a large FFT over 4 same OFDM symbols, which means a 4 x 64 FFT, which should give me 48 (corresponding to occupied tones) peaks instead of a band.
The theory of peaks can be illustrated simply by the FFT theory. I can share the reference for that if needed.
I hope I am clear here.
On Sun, Jan 3, 2016 at 7:06 PM, Martin Braun <address@hidden> wrote:
Typically, you'd do 1 (I)FFT per OFDM symbol. I recommend having a look
at the tx_ofdm.grc and rx_ofdm.grc examples, which show a full flow graph.
M
On 01/03/2016 01:22 PM, Aditya Virmani wrote:
> Hello everyone
>
> Iam trying to do run-time FFT over multiple OFDM symbols.
> For this, I feed in the same sequence of bytes for each OFDM symbol.
> As per my understanding, as I am feeding in the same sequence of bytes,
> it should be generating same set of symbols, and then further if we do
> FFT over multiple symbols which are same, we should get a sequence of
> peak-and-zeros, instead of continuous FFT graph.
> This works fine if I do just with two symbols (screenshot of that
> attached), but when I feed in more than two, it results in error ( I
> think that the data doesn't reach the FFT block concurrently, because
> the graph I get is same when I feed Null source and a Symbol source to
> FFT).
>
> It would be great if someone can guide me with it.
>
>
> Regards
> Aditya Virmani
>
>
> _______________________________________________
>
_______________________________________________
Discuss-gnuradio mailing list
address@hidden
https://lists.gnu.org/mailman/listinfo/discuss-gnuradio
