ANOMALY DETECTION IN AERONAUTICAL NETWORKS USING MARKOV CHAINS

Abstract

This work presents a theoretical proposal for anomaly detection in aeronautical communication networks using Markov Chains. The objective was to model the expected behavior of communication systems, such as those used during different flight phases, in order to identify deviations that may represent failures or attack attempts. The methodology consisted of defining states that represent the operational stages of a flight, building probabilistic transition matrices, and analyzing state sequences, considering improbable or non-existent transitions as anomalies. The results obtained through simulations and iterative calculations demonstrated that the model is able to identify anomalous patterns in an interpretable way, even in scenarios with restricted access to real data. It was found that the use of Markov Chains provides a simple, efficient, and applicable approach to anomaly detection in aeronautical networks, standing out for its clarity in representing states and for its ability to highlight critical events. It is concluded that the application of this technique contributes to strengthening cybersecurity in aviation and to preventing incidents in critical infrastructures.