MATLAB Script for BER vs. SNR for M-QAM, M-PSK, QPSk, BPSK
Output
.png)
If M>8, the distance between constellation points is short, and a higher Eb/No (SNR per Bit) ratio is required to reach the desired BER. Although the mapping from the data bits is arbitrary, some data bits are typically used. Every constellation point in the M-PSK constellation has two neighbors, each with an equal chance of making an error. As there are four bits per symbol, BER assumes a one-bit error for every mistake in a character. The demodulator in PSK must be able to calculate the received sinusoid's phase about some reference phase. While using the same bandwidths as ASK, PSK is less prone to errors than ASK. Also, using bandwidth with a significant data rate is more effective.
In the above figure, it is clear that PSK is more robust than QAM in the context of noise resilience. QAM modulations, including 16-QAM, are sensitive to both amplitude and phase errors. As you increase the number of constellation points (e.g., from 16-QAM to 8-PSK), the signal becomes more susceptible to amplitude and phase noise. In contrast, PSK modulations primarily rely on phase information and may be less sensitive to amplitude variations. This can make 8-PSK more robust in this scenario.
QAM schemes require a higher SNR to achieve the same error rates as PSK schemes with the same number of constellation points. This means that 16-QAM may require a higher SNR than 8-PSK to achieve a satisfactory bit error rate (BER) or symbol error rate (SER). In practical communication systems, achieving the necessary SNR can be challenging, especially in noisy or fading channels.
MATLAB Code for BER vs SNR for m-ary QAM
clc;
clear all;
close all;
% Set parameters
snr_dB = -20:2:20; % SNR values in dB
qam_orders = [4, 16, 64, 256]; % QAM modulation orders
% Loop through each QAM order
for qam_order = qam_orders
% Calculate theoretical BER using berawgn
ber = berawgn(snr_dB, 'qam', qam_order);
% Plot the results
semilogy(snr_dB, ber, 'o-', 'DisplayName', sprintf('%d-QAM', qam_order));
hold on;
end
% Add labels and legend
title('BER vs SNR for Variable QAM');
xlabel('SNR (dB)');
ylabel('Bit Error Rate (BER)');
grid on;
legend('Location', 'best');
clear all;
close all;
% Set parameters
snr_dB = -20:2:20; % SNR values in dB
qam_orders = [4, 16, 64, 256]; % QAM modulation orders
% Loop through each QAM order
for qam_order = qam_orders
% Calculate theoretical BER using berawgn
ber = berawgn(snr_dB, 'qam', qam_order);
% Plot the results
semilogy(snr_dB, ber, 'o-', 'DisplayName', sprintf('%d-QAM', qam_order));
hold on;
end
% Add labels and legend
title('BER vs SNR for Variable QAM');
xlabel('SNR (dB)');
ylabel('Bit Error Rate (BER)');
grid on;
legend('Location', 'best');
Output
Fig: BER vs SNR graph for Various QAM
Copy the aforementioned MATLAB Code from Here
MATLAB Code for BER vs SNR for m-ary PSK
clc;
clear all;
close all;
% Parameters
num_symbols = 1e5; % Number of symbols
snr_db = 0:2:20; % Range of SNR values in dB
% PSK orders to be tested
psk_orders = [2, 4, 8, 16, 32];
% Initialize BER arrays
ber_results = zeros(length(psk_orders), length(snr_db));
% BER calculation for each PSK order and SNR value
for i = 1:length(psk_orders)
psk_order = psk_orders(i);
for j = 1:length(snr_db)
% Generate random symbols
data_symbols = randi([0, psk_order-1], 1, num_symbols);
% Modulate symbols to generate signal
modulated_signal = pskmod(data_symbols, psk_order);
% Add AWGN to the signal
snr_linear = 10^(snr_db(j)/10);
received_signal = awgn(modulated_signal, snr_db(j), 'measured');
% Demodulate received signal
demodulated_symbols = pskdemod(received_signal, psk_order);
% Calculate BER
ber_results(i, j) = sum(data_symbols ~= demodulated_symbols) / num_symbols;
end
end
% Plot BER vs. SNR
figure;
semilogy(snr_db, ber_results(1, :), 'o-', 'DisplayName', 'BPSK');
hold on;
for i = 2:length(psk_orders)
semilogy(snr_db, ber_results(i, :), 'o-', 'DisplayName', sprintf('%d-PSK', psk_orders(i)));
end
title('BER vs. SNR for Various PSK Schemes');
xlabel('SNR (dB)');
ylabel('Bit Error Rate (BER)');
legend('Location', 'best');
grid on;
hold off;
clear all;
close all;
% Parameters
num_symbols = 1e5; % Number of symbols
snr_db = 0:2:20; % Range of SNR values in dB
% PSK orders to be tested
psk_orders = [2, 4, 8, 16, 32];
% Initialize BER arrays
ber_results = zeros(length(psk_orders), length(snr_db));
% BER calculation for each PSK order and SNR value
for i = 1:length(psk_orders)
psk_order = psk_orders(i);
for j = 1:length(snr_db)
% Generate random symbols
data_symbols = randi([0, psk_order-1], 1, num_symbols);
% Modulate symbols to generate signal
modulated_signal = pskmod(data_symbols, psk_order);
% Add AWGN to the signal
snr_linear = 10^(snr_db(j)/10);
received_signal = awgn(modulated_signal, snr_db(j), 'measured');
% Demodulate received signal
demodulated_symbols = pskdemod(received_signal, psk_order);
% Calculate BER
ber_results(i, j) = sum(data_symbols ~= demodulated_symbols) / num_symbols;
end
end
% Plot BER vs. SNR
figure;
semilogy(snr_db, ber_results(1, :), 'o-', 'DisplayName', 'BPSK');
hold on;
for i = 2:length(psk_orders)
semilogy(snr_db, ber_results(i, :), 'o-', 'DisplayName', sprintf('%d-PSK', psk_orders(i)));
end
title('BER vs. SNR for Various PSK Schemes');
xlabel('SNR (dB)');
ylabel('Bit Error Rate (BER)');
legend('Location', 'best');
grid on;
hold off;
Output
Fig: BER vs SNR graph for various PSK
Copy the above code from here
Read more about