Abstract
Nowadays, ensuring the security of wireless networks has become a serious problem because of the constant emergence of various cyber threats, which in turn require effective mechanisms for their detection and prevention. Bayesian graphs have shown notable advantages, particularly due to their capability to reflect complex interdependencies, adapt to dynamically changing conditions, and provide predictive insights for risk management. In this research, a Bayesian graph-based approach is proposed for systematically structuring and analyzing security states within wireless networks. The practical applicability of the suggested model was assessed through simulations involving common network attacks, specifically man-in-the-middle, denial of service, and node substitution scenarios. The obtained out-comes highlight the efficiency of Bayesian graphs in determining vulnerabilities and shaping proactive defensive measures. The presented method is particularly relevant in the context of smart city infrastructure, offering integration possibilities with other artificial intelligence tools, as well as potential applications across broader cybersecurity domains.