Название: Attack-and-Defense Games for Control Systems
Автор: Huanhuan Yuan, Yuan Yuan, Huapeng Dong, Yuanqing Xia
Издательство: CRC Press
Год: 2024
Страниц: 252
Язык: английский
Формат: pdf (true)
Размер: 16.0 MB

Complex control systems, including cyber-physical and cloud control systems, are in open network environments and are often confronted with threats from cyberspace, physical space and even cloud service. With diversified and intelligent attack patterns and improvements in attack capabilities, non-contact damage can be widespread. In this book, the authors use a formal, mathematical approach to introduce their recent research findings to describe and design attack and defense strategies using game theoretic method. The book is divided into three sections, focusing on strategies for resilience against deception attacks and DoS attacks, and protecting cloud control systems against threats. In these sections, the authors address topics such as secure and distributed filtering, attack detection and disturbance rejection, resilient state estimation, and resilient control, and techniques such as Stackelberg games, hierarchical games, and active eavesdropping. Through this book readers will be able to design effective defense strategies for complex control system to achieve resilience for closed-control cyber physical systems, network and cloud systems.

This chapter considers a cyber-physical system with multiple remote state estimation subsystems under denial-of-service (DoS) attack. Suppose that there are multiple distributed estimation systems, in each of which a sensor monitors the system and sends its local estimation to a remote estimator over one of multiple wireless channels. A DoS attacker emits noise power to jam the wireless channels. A scheduler is installed to dispatch each sensor to transmit information through a specific channel to minimize the total estimation error covariance on account of energy-saving, whereas, the DoS attacker attempts to jam a channel to realize an opposite objective. With considering interference from other individuals, a multi-sensor multi-channel remote state estimation model is constructed. A myopic policy and a long-term online interaction strategy under a varying network environment are investigated and compared by solving a two-player zero-sum game and a Markov game, respectively.