Investigating Unexpected Power Increase with Analogue FDM

[thavamaran]

Hi guys, I am trying to implement Analogue FDM on top of my single carrier modulation (eg: QAM and QPSK) of Multi carrier modulation (eg: OFDM). For an example, I have two different pair of I and Q channels, where first pair is upconverted at 2.2 GHz and the next pair at say 2.7 GHz and them MUX them together as in FDM.

I model them in matlab. Then when I downconvert them, it is possible to downconvert them without Bandpass filter as its ideal case in matlab, with using correlator receiver and correlate them at exact upconverted frequency to remove the frequency component.

This all works perfectly fine, the problem is, when I transmit a signal say at 2.2 GHz without any MUX, I get SNR of 16 dB at the BER that I want, but when I MUX two different carriers and downconvert them, my SNR reaches almost 22 dB at the same BER.

Im really curious to know what would be the problem, why it needs higher energy, is there anyway that I can solve this 6 dB of extra power. Please advice.

 

[tech99]

I think you will find that it is just the difference between PSK and 4-QAM. You are putting twice as many bits into the spectrum.