I can guide you on the steps and what functions could be used.

You would have to build custom functions for the GFDM transmitter and receiver parts since MATLAB doesn't provide built-in functions specifically for GFDM. Typically, you might start with baseband modulation of the data symbols, map these onto GFDM subcarriers, apply the necessary pulse shaping, and then sum these up to generate the GFDM signal.

Please note that you'll need a good understanding of the GFDM and OFDM signal processing chain to implement and compare these systems in MATLAB. I recommend checking out the MATLAB documentation and various online resources to get started with this project. If you have specific questions or need help with specific parts of this project, feel free to ask!

clear all

close all

clc

%% Define simulation parameters

N = 1024; % Number of subcarriers

T = 1e-6; % Symbol time

Fs = N/T; % Sampling frequency

SNRdB = 10:2:30; % Range of SNR values to simulate

%% Generate random binary data

data = randi([0,1],1,N);

%% OFDM modulation

% Perform IFFT on data

x = ifft(data);

% Add cyclic prefix

cp_length = round(N/4);

x_cp = [x(end-cp_length+1:end), x];

% Serialize data

s = reshape(x_cp,1,[]);

%% GFDM modulation

% Generate random symbols

symbols = qammod(randi([0,3],1,N),4);

% Define prototype filter

h = root_raised_cosine(T,Fs,'FilterSpanInSymbols',6,'RaisedCosineBeta',0.5);

% Apply pulse shaping

gfdm_signal = gfdm(symbols,h,T,Fs,'NumSubcarriers',N);

%% Simulate transmission over AWGN channel

for i = 1:length(SNRdB)

% Convert SNR from dB to linear scale

SNR = 10^(SNRdB(i)/10);

% Compute noise variance

noise_var = 1/SNR;

% Add noise to signals

rx_ofdm = awgn(s,SNRdB(i),'measured');

rx_gfdm = awgn(gfdm_signal,SNRdB(i),'measured');

% OFDM demodulation

rx_ofdm = reshape(rx_ofdm,length(s)/N,N+cp_length);

rx_ofdm(:,1:cp_length) = [];

rx_ofdm_data = fft(rx_ofdm,[],2);

rx_ofdm_data = reshape(rx_ofdm_data,1,[]);

% GFDM demodulation

rx_gfdm_data = gfdm_demod(rx_gfdm,h,T,Fs,'NumSubcarriers',N);

% Compute BER

[ber_ofdm(i),~] = biterr(data,rx_ofdm_data);

[ber_gfdm(i),~] = biterr(data,rx_gfdm_data);

% Compute PAPR

papr_ofdm(i) = 10*log10(max(abs(x_cp).^2)/mean(abs(x_cp).^2));

papr_gfdm(i) = 10*log10(max(abs(gfdm_signal).^2)/mean(abs(gfdm_signal).^2));

% Compute complexity

comp_ofdm(i) = N;

comp_gfdm(i) = N*6; % Assuming pulse shaping filter length of 6

end

%% Plot results

figure;

semilogy(SNRdB,ber_ofdm,'b--',SNRdB,ber_gfdm,'r-','LineWidth',2);

xlabel('SNR (dB)');

ylabel('BER');

legend('OFDM','GFDM');

figure;

plot(SNRdB,papr_ofdm,'b--',SNRdB,papr_gfdm,'r-','LineWidth',2);

xlabel('SNR (dB)');

ylabel('PAPR (dB)');

legend('OFDM','GFDM');

figure;

plot(SNRdB,comp_ofdm,'b--',SNRdB,comp_gfdm,'r-','LineWidth',2);

xlabel('SNR (dB)');

ylabel('Complexity');

legend('OFDM','GFDM');