Carrier Frequency Offset (CFO) Impairment Simulation in MATLAB

This section demonstrates Carrier Frequency Offset (CFO) impairment and its effect on the constellation diagram using MATLAB code. Parts A and C of the MATLAB code are assumed to be identical to those described on the AWGN page (referenced elsewhere, presumably).

MATLAB Source Code

The core of the CFO implementation is shown in Part B:

% PART B
foff=input('Enter Frequency offset:');%= 10e3; % Hz
foff=(foff)/20e6;   % 20MHz is the Bandwidth of the system
map_out_foff=frequency_offset(mapper_out,foff);
figure;plot(real(map_out_foff),imag(map_out_foff),'r+');title('constellation with Frequency offset');

This code snippet first prompts the user to enter the frequency offset in Hz. This offset is then normalized by the system bandwidth (20MHz in this example). The frequency_offset function (presumably a user-defined function not shown here) applies the calculated frequency offset to the mapper_out signal. Finally, the resulting constellation diagram with the applied CFO is plotted.

Input and Output Constellation Diagrams

The following images illustrate the constellation diagram before and after the introduction of Carrier Frequency Offset.

Input constellation diagram.

Constellation diagram after applying Carrier Frequency Offset.

As you can see, the CFO causes a rotation of the constellation points, which can significantly degrade the performance of the receiver if not properly compensated for. The severity of the rotation is directly proportional to the magnitude of the frequency offset.