$\%\%$ Parameters

N $=5$; $\%$ Number of users

L $=5$; $\%$ Number of subcarriers

B $=3$; $\%$ Near users

phi $=2$; $\%$ Far users

alpha $=$ randi$([0,1],$ L$,$ N$)$; $\%$ Power allocation for each user and subcarrier

PT $=10$; $\%$ Transmit power

n $=2$; $\%$ User of interest

hn $=$ randn$($L$,1)$; $\%$ Channel coefficients for each subcarrier

sigma$2=0\mathrm{.}01$; $\%$ Noise variance

S $=$ randi$([0,1],$ L$,$ N$)$; $\%$ Transmitted symbols for each user and subcarrier

g $=$ ones$($N$,1)$; $\%$ Scaling factor

wn $=$ sqrt$($sigma$2)*$ randn$($L$,1)$; $\%$ AWGN noise

sik $=$ randi$([0,1],$ N$,$ L$)$; $\%$ Modulated symbols for each user and subcarrier

$\%$ Preallocate for received signals

yn$\_$scalar$\_$C$\_$NOMA $=$ zeros$($L$,1)$;

yn$\_$scalar$\_$IM$\_$NOMA$\_$CR $=$ zeros$($L$,1)$;

$\%\%$ Loop through subcarriers: C$-$NOMA Received Signal Calculation

for k $=1$:L

$\%$ Intended signal for the nth user on the k$-$th subcarrier

intended$\_$signal $=$ hn$($k$)*$ sqrt$($alpha$($k$,$ n$)*$ PT$)*$ sik$($n$,$ k$)$;

$\%$ Interference signals from other users

interference$\_$signal $=0$;

for i $=1$:N

if i ~$=$ n

interference$\_$signal $=$ interference$\_$signal $+...$

hn$($k$)*$ sqrt$($alpha$($k$,$ i$)*$ PT$)*$ sik$($i$,$ k$)$;

end

end

$\%$ Received signal with interference and noise $($C$-$NOMA$)$

yn$\_$scalar$\_$C$\_$NOMA$($k$)=$ intended$\_$signal $+$ interference$\_$signal $+$ wn$($k$)$;

end

$\%\%$ IM$-$NOMA$-$CR Received Signal Calculation

for k $=1$:L

$\%$ Calculating x$\_$k as per the equation

xk $=$ sum$($sqrt$($alpha$($k$,1$:B$).\text{'}*$ PT$).*$ sik$(1$:B$,$ k$))+...$

sum$($sqrt$($alpha$($k$,$ B$+1$:N$-$phi$).\text{'}*$ PT$).*$ sik$($B$+1$:N$-$phi, k$))$;

$\%$ Calculating phi$\_$k as per the equation

phik $=$ sum$($sqrt$($alpha$($k$,1$:B$).\text{'}*$ PT$).*$ sik$(1$:B$,$ k$))+...$

sum$($sqrt$($alpha$($k$,$ B$+1$:N$-$phi$).\text{'}*$ PT$).*$ sik$($B$+1$:N$-$phi, k$))+...$

exp$(1$j $*$ pi $/2)*$ sum$($sqrt$($alpha$($k$,$ N$-$phi$+1$:N$).\text{'}*$ PT$).*$ sik$($N$-$phi$+1$:N$,$ k$))$;

$\%$ Received signal with noise $($IM$-$NOMA$-$CR$)$

yn$\_$scalar$\_$IM$\_$NOMA$\_$CR$($k$)=$ hn$($k$)*$ xk $+$ wn$($k$)$;

end

$\%\%$ Maximum Likelihood $($ML$)$ Detection for IM$-$NOMA$-$CR

possible$\_$symbols $=[0,1]$; $\%$ BPSK Symbols $\{0,1\}$

detected$\_$symbols$\_$ml$\_$IM$\_$NOMA$\_$CR $=$ zeros$($L$,1)$;

for k $=1$:L

min$\_$distance $=$ Inf; $\%$ Initialize with a high value

best$\_$symbol $=0$;

$\%$ Search over all possible transmitted symbols

for x$\_$candidate $=$ possible$\_$symbols

$\%$ Calculate Euclidean distance for each candidate

distance $=$ abs$($yn$\_$scalar$\_$IM$\_$NOMA$\_$CR$($k$)-$ hn$($k$)*$ x$\_$candidate$)^2$;

$\%$ Find the symbol with minimum distance

if distance $$ min$\_$distance

min$\_$distance $=$ distance;

best$\_$symbol $=$ x$\_$candidate;

end

end

$\%$ Assign detected symbol

detected$\_$symbols$\_$ml$\_$IM$\_$NOMA$\_$CR$($k$)=$ best$\_$symbol;

end

$\%$ Display the detected symbols for IM$-$NOMA$-$CR

disp$(\text{'}$ML Detected Symbols for IM$-$NOMA$-$CR:$\text{'})$;

disp$($detected$\_$symbols$\_$ml$\_$IM$\_$NOMA$\_$CR$)$;

$\%\%$ SNR vs BER Calculation

SNR$\_$dB $=0$:$5$:$50$; $\%$ SNR values in dB

SNR$\_$linear $=10.^($SNR$\_$dB $/10)$; $\%$ Convert SNR to linear scale

$\%$ Calculate BER trends $($replace with real logic if needed$)$

BER$\_$C$\_$NOMA $=0\mathrm{.}5./$ SNR$\_$linear; $\%$ C$-$NOMA BER trend

BER$\_$IM$\_$NOMA$\_$CR $=0\mathrm{.}8./$ SNR$\_$linear; $\%$ IM$-$NOMA$-$CR BER trend

$\%\%$ Plotting the BER Comparison Graph

figure;

semilogy$($SNR$\_$dB$,$ BER$\_$C$\_$NOMA, $\text{'}-$g$^\text{'},$ 'LineWidth', $1\mathrm{.}5,$ 'DisplayName', $\text{'}$C$-$NOMA ML All'$)$; hold on;

semilogy$($SNR$\_$dB$,$ BER$\_$C$\_$NOMA $*1\mathrm{.}1,\text{'}-$ks$\text{'},$ 'LineWidth', $1\mathrm{.}5,$ 'DisplayName', $\text{'}$C$-$NOMA ML Near'$)$;

semilogy$($SNR$\_$dB$,$ BER$\_$C$\_$NOMA $*0\mathrm{.}9,\text{'}-$bo$\text{'},$ 'LineWidth', $1\mathrm{.}5,$ 'DisplayName', $\text{'}$C$-$NOMA ML Far'$)$;

semilogy$($SNR$\_$dB$,$ BER$\_$IM$\_$NOMA$\_$CR$,\text{'}-$r$*\text{'},$ 'LineWidth', $1\mathrm{.}5,$ 'DisplayName', $\text{'}$IM$-$NOMA$-$CR ML All'$)$;

semilogy$($SNR$\_$dB$,$ BER$\_$IM$\_$NOMA$\_$CR $*1\mathrm{.}1,\text{'}-$md$\text{'},$ 'LineWidth', $1\mathrm{.}5,$ 'DisplayName', $\text{'}$IM$-$NOMA$-$CR ML Near'$)$;

semilogy$($SNR$\_$dB$,$ BER$\_$IM$\_$NOMA$\_$CR $*0\mathrm{.}9,\text{'}-$cp$\text{'},$ 'LineWidth', $1\mathrm{.}5,$ 'DisplayName', $\text{'}$IM$-$NOMA$-$CR ML Far'$)$;

hold off;

$\%$ Set axis labels and title

xlabel$(\text{'}$SNR $($dB$)\text{'})$;

ylabel$(\text{'}$BER$\text{'})$;

title$(\text{'}$BER Comparison between C$-$NOMA and IM$-$NOMA$-$CR$\text{'})$;

legend$(\text{'}$Location$\text{'},$ 'southwest'$)$;

grid on;

In this i have used ML detector Now I want an matlab code using log likelihood detector and maximum ratio combining Detector.output should be the fig$-4.$ you can use any AI tools but i want same graph as output.but graphs are different i will report.You can use any AI tools but graph is important

Fig. $4.$ BER comparison between C$-$NOMA and IM$-$NOMA$-$CR$.$