After taking IFFT at transmitter for OFDM, I want to plot the out-of-band emission to compare between OFDM and SC-FDMA (using MATLAB)
I tried using the 'periodogram' function in MATLAB by plotting the psd by [pxx,f] = periodogram(ofdm_Tx); plot(f/pi,10*log10(pxx))
But it seems it's giving very noisy type of figure. Is there any effective way to observe the out-of-band emission?
'out of band' is best assessed with a comprehensive (complete and robust) model that includes 'good enough' modelling of whatever is above and below the frequency band of interest
or
do real measurments, let me explain:
1. You have not supplied a model, in Simulink or MATLAB that simulates signals, noise and interence outside the frequencies you have generated with a couple of code lines.
Therefore it is reasonable to assume that you do not have the model.
For instance, do you have the 5G toolbox?
because using the right toolboxes brings you into the game of attempting to simulate outside of the strict parameters of your system.
Why would we be simulating after all, if not not to try seeking data, not where we obviously know it all, but where we don't know yet?
2. The 'noisy' 'out of band' of band you get is correct.
But you are not ampilfying your signal, so you have left your OFDM signal to fence with similar amplitude to noise.
OFDM is a really robust modulation, so even below noise, or even on same band of other stronger signals, your OFDM signal may deliver.
Amplify : No way around, to push down nearby noise and see the shoulders that define your OFDM signal.
3. So, out of experience, if you have the means to do it, build the real system and directly measure shoulders, out of band, and whatever needed to be measured.
And then you may try building a model that fits real measurments.