Hi Robert,
As per my understanding, you would like to convolute a baseband complex signal with the baseband equivalent channel response based on RF-measured s2p parameters.
The function 'impulse(rat_fit, 50e-9,100)' doesn't yield a complex response because the 'impulse' function is typically used for linear time-invariant (LTI) systems, and it returns the response of the system to an impulse input. For a rational model created from S-parameter measurements, the impulse response should be real-valued because it represents the response to a delta function. I would suggest you try to manually compute the impulse response from the frequency response using an inverse Fourier transform. If 'timeresp' doesn't accept complex signals, you could handle the real and imaginary parts separately using the 'conv' function.
The following example code snippet may help you achieve this:
% I have used a sample data for explaining the method.
S = sparameters('defaultbandpass.s2p');
freq = S.Frequencies;
data = rfparam(S,2,1);
rat_fit = rational(freq,data)
% Now rat_fit is the rational function object that models the S21 response
% Compute frequency response
[H, f] = freqresp(rat_fit, freq);
% Obtain impulse response by inverse Fourier transform (if necessary)
impulseResponse = ifft(H);
% Sample Baseband signal
fs = 100e6; % sample rate of 100 MHz
t = 0:1/fs:1e-6; % time vector for 1 microsecond duration
% Generate a complex exponential signal with a frequency offset
f_offset = 5e6; % frequency offset of 5 MHz
basebandSignal = exp(1i*2*pi*f_offset*t);
% Add some noise to the signal
noise = 0.1*(randn(size(t)) + 1i*randn(size(t)));
basebandSignal = basebandSignal + noise;
% Separate baseband signal into real and imaginary parts
realSignal = real(basebandSignal);
imagSignal = imag(basebandSignal);
% Convolve each part with the impulse response
convReal = conv(realSignal, impulseResponse, 'same');
convImag = conv(imagSignal, impulseResponse, 'same');
% Combine the convolved signal parts into a complex signal
convolvedSignal = convReal + 1i * convImag;
disp(convolvedSignal);
Kindly refer to the following documentation links to have more information on:
- 'ifft' function: https://www.mathworks.com/help/matlab/ref/ifft.html
- 'conv' function: https://www.mathworks.com/help/matlab/ref/conv.html
Hope that helps!
