The scripts differentiate between the Markov modeling and the simulation part.
The simulation contains the reference system and other evaluated configurations. "reference_system.py" generates the results for the reference system, while "run.py" generates the results for the parameter study. Each result is saved in a new directory with the corresponding parameter selections serving as the name.
The Markov modeling part contains four scripts. The 'm_gi_1.py' script contains the general Markov model and is required to plot Figure 4 and Figure 5. It is possible to adapt arrival rate lambda to achieve results for other systems. Then, 'DES_waiting_time.py' is required to generate Figure 6. The goal is to get a general dimensioning of the system where again, the target waiting time and the percentage services are finished within that time can be adapted. Afterwards, 'markov_window_eval.py' is used to generated Figure 8 for a dimensioning of receive windows to identify the target number of messages per channel achieving a waiting time below x for different probabilities. All values are adjustable. Finally, the 'energy_assessment.py' script is used to generate Figure 9 and assess energy wastage in the system using our modeling approach.