Is it possible to set channel type to gaussian?

Hi, I was wondering if there is an option to change the channel type from Rayleigh to Gaussian in LTE Toolbox.

I am simulating downlink channels and I would like to simulate first a channel with just AWGN.

I have seen ModelType can be GMEDS or Dent, both Rayleigh.

channel.ModelType = 'GMEDS';         % Rayleigh fading model type

How can I make the channel type to be Gaussian?

Thank You.

Answer 1

[Arthi Sathyamurthi]

As of now the available options for channel type in the channel model configuration are the Rayleigh fading model types such as ‘GMEDS’ and ‘Dent’. You can still switch off fading completely and implement a static MIMO channel model by setting the ‘DelayProfile’ input argument as ‘Off’. Here the matrix containing the link between each transmit and receive antennas will have zero delay and unit gain.

For more information you can refer to the input arguments subsection of the lteFadingChannel documentation page.

[PAU TAÑA VIÑETA]

Hi! thank you for your answer.

The thing is this way does not mean the channel model is Gaussian either - I suppose.

I tried setting rxWaveform = txWaveform, and then rxWaveform = rxWave form + noise, so no fadding channel is applied, just AWGN, but now I get BLER = 1 for all SNR points. Maybe it is because of some normalitzation, I don't know.

So, I assume there is no way to simulate Gaussian model, isn't it?

Thank you again.

P.

 % Pass data through channel model
             channel.SamplingRate = ofdmInfo.SamplingRate;
            [rxWaveform,fadingInfo] = lteFadingChannel(channel, txWaveform);
             rxWaveform=txWaveform; % <-------
             
            % Calculate noise gain including compensation for downlink power
            % allocation
            SNR = 10^(SNRdB(snrIdx)/20);

            % Normalize noise power to take account of sampling rate, which
            % is a function of the IFFT size used in OFDM modulation, and
            % the number of antennas
            N0 = 1/(sqrt(2.0*enb.NBRefP*double(ofdmInfo.Nfft))*SNR);
            
            % Create additive white Gaussian noise
            noise = N0*complex(randn(size(rxWaveform)), ...
                                randn(size(rxWaveform)));

            % Add AWGN to the received time domain waveform   
            rxWaveform = rxWaveform + noise;

[Arthi Sathyamurthi]

Hi Pau, since I don’t know the receiver blocks of your program the possible workaround/solution is to try checking your equations on normalizing noise power. In case you feel the equation is right, check if the AWGN noise generated is right (try plotting the PDF). Also one another workaround could be increasing the range of SNR.