Theoretical questions about RF baseband complex signals

Hi,

I am trying to perform some RF simulation with different types of signal through a non-linearity type saleh MemorylessNonlinearity.

For this model, Expected input to be complex.

I try to define these signals:

1 - a continuous wave signal and a two sinusoids signal.

I would define it in time domain but I admit I don't see how to define it as baseband complex signals.

2 - a band limited noise signal between Fc-BW/2 and Fc+BW/2 with Fc = central frequency, BW = bandwidth

I guess It's easier to define it in the frequency domain for the qspect band limited and after thanks to an inverse FFT = ifft to get the waveform signal.

I have tried this :

% Identify the locations of frequency bins. These will be used to zero out the elements of Xin that are not in the desired band

freqbins = (0:Nbre_sample)'/Nbre_sample*Fsample;

Xin=2*rand(Nbre_sample,1); % or Xin=exp(i*2*pi*rand(Nbre_sample,1));

Xin(find((freqbins < F_min)|(freqbins > F_max))) = 0;

% Xin = [Xin; conj(flipud(Xin(2:end-1)))] ;

xin = sqrt(Nbre_sample)*ifft(Xin);

But I have a doubt, do I have to create a symetric spectrum with line command : Xin = [Xin; conj(flipud(Xin(2:end-1)))]. Doing this I have xin real and no more complex.

the signal is described originally described in the frequency domain and around the carrier frequency. How do I return to baseband signals as input for the amplifier? taking the envelope?

Thanks for your time