Nash Game Modeling for Network Defense via Node Hiding and Resource Allocation

Modern network systems face persistent security challenges arising from asymmetric information, limited defensive resources, and attackers with diverse and often bounded rationality. These conditions make it difficult for defenders to maintain network connectivity while preventing adversaries from exploiting structural vulnerabilities. To address this challenge, this paper proposes a game-theoretic framework that integrates node hiding and resource allocation within a zero-sum Nash game formulation. In this model, the defender and attacker simultaneously optimize their strategies under constrained budgets, with the defender employing structural deception and selective reinforcement to preserve connectivity, and the attacker choosing targets based on partially observed or inferred topology. Three attacker types - Naive, Bayesian, and Random - are introduced to represent varying rationality levels and information utilization. Experimental results demonstrate that (i) node hiding effectively disrupts low-rationality attackers, (ii) resource allocation strengthens robustness against more rational adversaries, and (iii) their combination produces a synergistic improvement in overall resilience. Across all attacker models, the joint strategy consistently yields the highest defensive utility, confirming that coupling topological misdirection with targeted resource allocation provides an effective and generalizable defense mechanism.