First, it's true that the simulation takes a long time, so don't worry! To speed the simulation (but obtain less accurate results), you can decrease the number of blocks transmitted at each SNR level. Specifically, modify this line:
Second, there is a line in fairul's code where it seems to be loading results of non-cooperative mode previously stored in a file. This file was not uploaded, so what we can do is to comment (or delete) the this line, which is:
Code:
load noCooperation
and of course remove any variables depending on the file (noCooperation.mat). Specifically, replace the following line: