/DTNRoutingMarkov

A finite-state discrete-time homogeneous Markov chain model for analysing DTN Routing in a Urban Sensing Scenario

Primary LanguagePythonGNU General Public License v3.0GPL-3.0

DTNRoutingMarkov

Python scripts developed for a dissertation titled Delay Tolerant Network Routing in the context of a Master in Informatics Engineering.

Includes modelling and analysis of a urban sensing scenario with a Delay Tolerant Network, based on a finite-state discrete-time homogeneous Markov chain model. The scripts are organized as follows:

  • markov_chain.py - Contains the model and its corresponding analysis.

  • communication_routing_models.py - Contains communication models and routing strategies (used in markov_chain.py)

  • mobility_models.py - Contains mobility models (used in markov_chain.py)

  • tpm_simulation.py - A simulation of this model, based on its Transition Probability Matrix (used for verification)

  • entity_simulation.py - A simulation of this model, based on the behaviour of its entities (used for verification)

All scripts were tested for Python 3.7 (check requirements.txt).