Abstract:Advanced persistent threat (APT) is a kind of stealthy, sophisticated, and long-term cyberattack that has brought severe financial losses and critical infrastructure damages. Existing works mainly focus on APT defense under stable network topologies, while the problem under time-varying dynamic networks (e.g., vehicular networks) remains unexplored, which motivates our work. Besides, the spatiotemporal dynamics in defense resources, complex attackers' lateral movement behaviors, and lack of timely defense make APT defense a challenging issue under time-varying networks. In this paper, we propose a novel game-theoretical APT defense approach to promote real-time and optimal defense strategy-making under both periodic time-varying and general time-varying environments. Specifically, we first model the interactions between attackers and defenders in an APT process as a dynamic APT repair game, and then formulate the APT damage minimization problem as the precise prevention and control (PPAC) problem. To derive the optimal defense strategy under both latency and defense resource constraints, we further devise an online optimal control-based mechanism integrated with two backtracking-forward algorithms to fastly derive the near-optimal solution of the PPAC problem in real time. Extensive experiments are carried out, and the results demonstrate that our proposed scheme can efficiently obtain optimal defense strategy in 54481 ms under seven attack-defense interactions with 9.64$\%$ resource occupancy in stimulated periodic time-varying and general time-varying networks. Besides, even under static networks, our proposed scheme still outperforms existing representative APT defense approaches in terms of service stability and defense resource utilization.

An adaptive attack model to network controllability

Attack Vulnerability of Network Controllability.

Adaptive Event-Driven Control for Continuous Positive Stochastic Jump Models with Cyber Attacks

A Framework of Hierarchical Attacks to Network Controllability

Optimal Repair Strategy Against Advanced Persistent Threats Under Time-Varying Networks

Resilient event-triggering adaptive neural network control for networked systems under mixed cyber attacks

Adaptive neural network tracking and controlling network attack node in WSN

New attack strategies for complex networks

Distributed Adaptive Consensus Control for Nonlinear Systems with Active-Defense Mechanism Against Denial-of-Service Attacks

Robustness of Non-Interdependent and Interdependent Networks Against Dependent and Adaptive Attacks

Robustness Of Controllability For Networks Based On Edge-Attack

Computationally efficient attack design for robustness analysis of air transportation networks

Finite‐time resilient control for networked control systems with multiple cyber‐attacks: Memory/adaptive event‐triggered scheme

Maximizing Network Resilience against Malicious Attacks

Probabilistic models for evaluating network edge's resistance against scan and foothold attack

Optimization for sequential communication line attack in interdependent power-communication network

Robustness of Interdependent Networks under Several Intentional Attack Strategies

ADP-based Remote Secure Control for Networked Control Systems under Unknown Nonlinear Attacks in Sensor and Actuator

Adaptive Neural Dissipative Control for Markovian Jump Cyber-Physical Systems Against Sensor and Actuator Attacks

Mitigation of Malicious Attacks on Networks

A Sampled-Data-Based Secure Control Approach for Networked Control Systems Under Random DoS Attacks