The confidence intervals are calculated via bootstrapping. This is performed in
+Results/SimulationsResults.m by calling the function
By default, a 95% confidence level is used when calling that function. You can find some details here: https://en.wikipedia.org/wiki/Confidence_interval .
Roughly speaking, they indicate the accuarcy of your simulated results. Large confidence intervals correspond to uncertain results, meaning that if you repeat the simulation, you are less likely to get a similar average value (of the throughput for example). In order to make them smaller, and therefore make the results more accurate and representative, you should increase the number of simulation samples by increasing the number of simulated frames, i.e., set
scStr.simulation.nFrames to a larger value in the scenario file.