Resilient Control for Cyber Physical Systems

Get Started. It's Free
or sign up with your email address
Rocket clouds
Resilient Control for Cyber Physical Systems by Mind Map: Resilient Control for Cyber Physical Systems

1. A survey on security control and attack detection for Industrial Cyber Physical systems (ding2018survey)

1.1. Presents an Overview of Recent Advances on Security Control and attack detection for ICPS

1.1.1. System models for Industrial CPS Discuss the purpose of using system models for understanding and analyzing system dynamics Unified system model before any analysis, due to tight coupling and high coordination between CPS, can be regarded as (dynamically reorganizing and re configuring control system with high degree of automation at multiple spatial and temporal dimensions), such a spatial temporal characteristic can be described as distributed parameter systems concerns - control and monitor are typically implemented on digital platforms, physical components run in a continuous way while cyber components are discrete, bridging the gap ultimately is inevitably challenging issue for modelling and verifying CPS this issue gives rise to a hybrid system model which can possible account for the complexity of a CPSs Distributed Parameter Systems Reviewed models and research Hybrid Systems Reviewed models and methods Markovian decision process Reviewed models and research

1.1.2. Three typical types of cyber attacks were discussed DOS, replay attacks and Deception attacks DOS attacks DOS attack is an attempt to make the system resources unavailable, attackers can fill buffers of user domains or kernel domains, jam the shared network medium to prevent devices from communicating or receiving or change the routing protocol Replay attacks A replay attack is a natural strategy, in which valid data transmission is maliciously or fraudulently repeated or delayed. Deception attacks A deception attack is a type of cyber attack, in which data integrity is modified for transmitted packets among different cyber parts, for example stuxnet worms which can reprogram the code running in a PLC or SCADA systems (a decption attack in different scenarios can also be know as false data injection attacks)

1.1.3. Performance analysis for Industrial CPSs / performance analysis is at an elementary stage especially in the theoretical level due to challenges from uncertainties in the environment, modelling errors of physical and software components stability research reviewed robustness recently a new notion of robustness is introduced to suppress the impact from both the bounded deviations of nominal behaviors and sporadic disturbances in finite steps and the satisfactory controller is designed under psuedo polynomila time security in comparison with conventional control systems, CPSs usually run a high risk of cyber attacks due to the introduction of communication networks and heterogeneous IT elements which give rise to security vulnerabilities, reliability/resilience it is worth pointing out that the requirements of robustness and security are commonly predetermined, it is impractical to design for all attacks, therefore this gives rise to resiliency, which refers to the ability of the system to recover after an attack

1.1.4. Attack Detection for ICPS/ cyber attacks are harmful therefore when they are detected and located in a timely fashion, the damage to the overall system can be controlled within a tolerable limit. Attack detection plays a crucial role in maintaining the performance of CPSs Two schemes to defend against cyber attacks Protect the important system components beforehand Identify false data injected by attackers afterwards

1.1.5. Security Control and Estimation for Industrial CPSs - important role in better understanding system dynamical behaviors and ececuting specific control tasks The maximum tolerant number of attacks is discussed. It is a highly related to a concept of strong observability under which the system states can be accurately reconstructed via attack resilient estimators. It is worth noting that a similar notion named s-saprse observability can be found. An lo-norm based state estimator is adopted to guarantee the boundedness of estimation errors caused by noises, modeling errors as well as cyber attacks Security control techniques for CPSs remain at an infant stage and few results can be found in literature. Networked control systems subject to cyber attacks have some preliminary results research in the context of CPS security a trade off between security and stability has also been receiving attention

1.1.6. Conclusions and Challenging issues System modelling and methodologies Attack Detection and compensation The system performance and the quality of service

2. A survey on the security of cyber-physical systems (wu2016e)

2.1. Section 2- A stealthy attack scheme close to covert attack that a malicious agent can use to successfully realize the attack without being detected

2.2. Section3 - Resilient defense strategy that a defender can use to quickly recover the nominal behavior of the NCS

2.3. Section 4 - Results