面向 DoS 攻击下安全状态估计的无线信息物理系统优化调度研究

Recently, the security issue of cyber-physical systems (CPS) has attracted much attention from different communities. In CPS, cyber threats emerge in many ways which can cause significant destruction to the system operation. The study of state estimation and resource scheduling of the system with multiple sensors and multiple channels under denial of service (DoS) attack are interesting and meaningful aspects in CPS. This proposal digs into the coupling relationship between multi-channel network and control subsystems in CPS, and focuses on the study of network resource scheduling against state estimation under DoS attack in wireless multi-channel CPS. The main research topics are: 1) based on the integrated model under DoS attack in wireless multi-channel CPS, analyzing the performance of remote state estimation against network resource constraints, and proposing the optimal remote state estimation design methods; 2) analyzing the network resource scheduling problem in wireless multi-channel CPS under DoS attack, designing the optimal resource scheduling strategy, and investigating the system stability problem under the designed scheduling; 3) providing the corresponding undetected optimal resource scheduling policy when a special attack detection mechanism is equipped at the remote estimator. The proposed research will promote the integration of control theories and wireless networking technologies, and provide basic theories and design guidelines for practical CPS applications.

近年来，信息物理系统(CPS)的安全问题引起了社会各界的广泛关注。在CPS中，网络威胁以多种方式出现并对系统运行造成重大破坏。对于多传感器多信道系统中存在拒绝服务(DoS)攻击干扰的情况，研究系统的状态估计和资源调度是CPS中的一项十分有意义且急需进行的工作。本项目深入挖掘CPS中网络与控制的耦合关系，重点研究无线多信道CPS中DoS攻击下针对状态估计的网络资源调度问题。主要研究内容如下：1）基于无线多信道CPS中针对DoS攻击的一体化模型，分析针对网络资源约束下远程状态估计性能，提出最优远程状态估计设计方法；2）分析DoS攻击下网络资源调度问题，设计最优资源调度策略，并考察该调度下系统稳定性问题；3）当远程估计器内置一个特殊的攻击探测机制时，提供相应的不被检测到的最优资源调度策略。本项目的研究有助于推动控制理论与无线网络技术的进一步融合，为信息物理系统面向实际应用提供基础理论和方法指导。

本项目主要围绕信息物理系统（CPS）这一交叉研究热点领域，深入挖掘信息物理系统中网络与控制的耦合关系，重点研究CPS中复杂动态信道下针对网络攻击的最优攻击调度和传感器能量受限情况下的网络资源分配等问题。项目的主要研究内容如下：.（1）结合马尔科夫等随机分布模型，构建了CPS中复杂动态信道下针对网络资源约束的攻击行为模型，进一步研究了无线信息物理系统中传感器在拒绝服务(DoS)攻击下的稳定状态估计和最优攻击调度问题；.（2）为了改善网络节点的能量受限问题，研究了当传感器可以接入能量采集器，可以从外部环境收集能量时，无线网络物理系统中采用能量获取技术的远程状态估计传感器的最优传输功率分配问题。.本项目的研究有助于推动控制理论与无线网络技术的进一步融合，为信息物理系统面向实际应用提供基础理论和方法指导。