0
$\begingroup$

I am studying the effect of time offset in OFDM system, I am interested in calculating the power of the ICI on each subcarrier, I am a bit confused about how to do it in Matlab, so how can I measure the power of ICI on a specific subcarrier in Matlab?

$\endgroup$
10
  • $\begingroup$ Do you understand what time offset does to each sub-carrier? $\endgroup$ – Dan Boschen Feb 4 '20 at 3:25
  • $\begingroup$ Yes, so basically I am studying the effect of timing jitter in OFDM systems, I understand it causes both a phase rotation and an ICI, I am interested in measuring the ICI power (on a specific subcarrier) vs. the timing offset. I could not find a way to measure ICI on a specific subcarrier using Matlab? $\endgroup$ – Amro Goneim Feb 4 '20 at 3:28
  • $\begingroup$ A delay in time is a linear phase in frequency-- so the phase rotation on each subcarrier gets successively larger (does that make sense?). What modulation is your sub-carrier? $\endgroup$ – Dan Boschen Feb 4 '20 at 3:40
  • $\begingroup$ Thanks a lot, I am using 4 and 16 QAM. I am aware of the effect of the phase rotation, however I could not find a proper way to measure the ICI on each subcarrier. I tried to to the following: in order to measure the ICI on subcarrier 1 (ICI effect that is caused by all other subcarriers on SC1) I send zero in subcarrier 1 (send nothing) and then the result of fft at the receiver for SC1 should be the ICI. However I did not get resonable results. Any help? $\endgroup$ – Amro Goneim Feb 4 '20 at 3:46
  • $\begingroup$ I'm not providing an answer below as I haven't done this specifically so there may be more direct approaches but I would proceed exactly as you suggested. Can you provide details in your question and why you don't think the results are not reasonable? I also had assumed that CFO was a more dominant source of ICI and still trying to think through how a static timing offset would cause any significant ICI $\endgroup$ – Dan Boschen Feb 4 '20 at 3:50
0
$\begingroup$

In order to measure the ICI power on a subcarrier x (ICI that is caused by all other subcarriers). Simply, consider an ideal system without any source of noise, keep all subcarriers turned on (activated) and turn off subcarrier x (through power loading or bit loading) so that subcarrier x does not convey any data. At the receiver, after FFT, the power on subcarrier x is the ICI power. Moreover, the measured EVM (in dB) at subcarrier x is indeed the ICI power in dB.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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