This article is concerned with the quantized control problem for neural networks with adaptive event‐triggered scheme (AETS) and complex cyber‐attacks. By fully considering the characteristics of ...cyber‐attacks, a mathematical model of complex cyber‐attacks, which consists of replay attacks, deception attacks, and denial‐of‐service (DoS) attacks, is firstly built for neural networks. For the sake of relieving the pressure under limited communication resources, an AETS and a quantization mechanism are employed in this article. By utilizing Lyapunov stability theory, adequate conditions ensuring the stability of neural networks are obtained. Moreover, the controller gain is derived by solving a set of linear matrix inequalities. At last, the usefulness of the proposed method is verified by a numerical example.
Industry 5.0 is a emerging transformative model that aims to develop a hyperconnected, automated, and data-driven industrial ecosystem. This digital transformation will boost productivity and ...efficiency throughout the production process but will be more prone to new sophisticated cyber-attacks. Deep learning-based Intrusion Detection Systems (IDS) have the potential to recognize intrusions with high accuracy. However, these models are complex and are treated as a black box by developers and security analysts due to the inability to interpret the decisions made by these models. Motivated by the challenges, this paper presents an explainable and resilient IDS for Industry 5.0. The proposed IDS is designed by combining bidirectional long short-term memory networks (BiLSTM), a bidirectional-gated recurrent unit (Bi-GRU), fully connected layers and a softmax classifier to enhance the intrusion detection process in Industry 5.0. We employ the SHapley Additive exPlanations (SHAP) mechanism to interpret and understand the features that contributed the most in the decision of the proposed cyber-resilient IDS. The evaluation of the proposed model using the explainability can ensure that the model is working as expected. The experimental results based on the CICDDoS2019 dataset confirms the superiority of the proposed IDS over some recent approaches.
A new resilient distributed secondary control for AC microgrids is studied based on event-triggered mechanisms and trust-reputation evaluation methods. When distributed generators (DGs) in a ...microgrid are subject to attacks, their transmitted state information would be tampered and thus affect the dynamics of normal generators. In order to isolate possible attacks, two types of trust evaluation metrics with different attack indices and time scales are designed, by which the performance of neighboring DGs can be assessed for specific practical demands. Based on the trust values of each neighbor, a reputation-propagation method is introduced at triggered time instants to determine whether a DG is under attack by comprehensively incorporating the opinion of mutual neighbors. The dynamic updating law of the communication edge weights is utilized with the derived reputation values. Based on this, a distributed Zeno-free event-triggered control protocols for voltage/frequency restoration and active power sharing are proposed. Sufficient conditions for picking proper control parameters are given in the main theorem. Lastly, the simulations are conducted in MATLAB/SimPowerSystems for several scenarios to validate the effectiveness of the proposed algorithms.
This article focuses on the economic environmental resource management for islanded Microgrids. To optimize the competing social welfare and environmental impact objectives simultaneously, by using ...the linear weighted sum method, a switched distributed algorithm is proposed to assign the energy among multiple heterogeneous generation devices and loads. Considering the vulnerability of distributed algorithms to the denial of service (DoS) attacks, we study the impact of the frequency and duration of such attacks on the algorithm performance. To obtain the optimal operation even under DoS attacks, sufficient conditions are presented to ensure the exponential convergence of the algorithm. Meanwhile, an event-triggered communication strategy is designed to reduce communication resource among participants. Finally, the effectiveness of the algorithm is illustrated by several case studies. Note to Practitioners -Resource management problem is a key issue in Microgrids to improve the economics of operation, which can be addressed by distributed cooperative algorithms. As a typical cyber-physical system, the optimality and convergence of the distributed algorithms can be easily disrupted by various cyber-attacks, such as DoS attacks. The existing distributed algorithms are typically implemented under ideal communication environments. To overcome the limitation and accommodate various resources in the future Microgrids, a distributed initialization-free algorithm is proposed that is resilient to DoS attacks. Besides, practitioners can adjust the coefficients of the algorithm based on the real operating situations to guarantee the convergence. Our future work will focus on designing resistance mechanisms and considering more practical constraints.
This paper addresses the problem of decentralized event-triggered H∞ control for neural networks subject to limited network-bandwidth and cyber-attacks. In order to alleviate the network transmission ...burden, a decentralized event-triggered scheme is employed to determine whether the sensor measurements should be sent out or not. Each sensor can decide the transmitted sensor measurements locally according to the corresponding event-triggered condition. It is assumed that the network transmissions may be modified by the occurrence of the random cyber-attacks. A Bernoulli distributed variable is employed to reflect the success ration of the launched cyber-attacks. The Lyapunov method is employed to derive a sufficient condition such that the closed-loop system is asymptotically stable and achieves the prescribed H∞ level. Moreover, the desired H∞ controller gains are derived provided that the sufficient condition is satisfied. Finally, illustrative examples are utilized to show the usefulness of the obtained results.
Recursive filtering for nonlinear systems, one of the core technologies of modern industrial systems, is an ever-increasing research topic from the control and computer communities. Some challenges ...from communication scheduling, limited bandwidth as well as security vulnerability have to be seriously handled though the applications of communication technologies bring into some conveniences. As such, it is of utmost significance in theory and great importance in applications to establish engineering-feasible recursive filtering algorithms for networked nonlinear systems. This paper focuses on the development of this topic and provides an up-to-date survey of the existing nonlinear filtering techniques. The introduction of three classes of communication protocols is first presented in great detail, and then comprehensive reviews and summaries of the nonlinear recursive filtering problems with Gaussian/non-Gaussian noises are elaborated according to different strategies responding to nonlinear functions or noises. Particularly, the reviews are layout from the extended Kalman filtering, the unscented/cubature Kalman filtering, the set-membership filtering as well as the
filtering. Furthermore, several challenging issues are raised to stimulate further related theoretical research and practical applications in this field.
In this study, a bandwidth allocation-based distributed event-triggering load frequency control (LFC) has been developed for smart grids to deal with hybrid cyber-attacks, for example, ...denial-of-service (DoS) attacks and false data injection (FDI) attacks. Firstly, to prevent hybrid cyber-attacks from causing open-loop unstable operation of the LFC systems, we propose a distributed event-triggering communication (ETC) strategy. To attain the maximum usage of bandwidth, a dynamic bandwidth allocation mechanism is integrated with the ETC approach on the basis of resource availability and error between the current state and equilibrium state. This bandwidth reservation and allocation approach aim at detecting attacks and assigning bandwidth to the different channels of distribution networks. Then, by virtue of the Lyapunov approach, the exponential stability criteria are established. Further, the exclusion of Zeno behavior of the designed systems is proved during the control process. Finally, comprehensive case studies show that the proposed method can improve the utilization rate of the network resource.
The connected vehicles (CVs) technology mainly relies on the traffic cyber physical systems (T-CPS) to realize the interconnection between CVs. However, the vulnerability of open CVs environment ...gives illegal attackers an opportunity. Obviously, CVs are susceptible to unlawful cyber-attacks (CAs) from infiltrating malicious messages. Accordingly, underlying cybersecurity threats can cause information exchange errors and communication failures between CVs. For example, attackers can spread false messages that interfere with normal communication between CVs, which results in CVs receiving incorrect traffic information to affect their decisions and actions. In addition, attackers can also apply denial of service (DoS) attack to overload the communication network, causing communication failures and network congestion between CVs. Therefore, a new coupled map car-following model is established by integrating the safety control against CAs on the global information (called for CAGI-CM model) under CVs platoon environment. In terms of cybernetics, the necessary conditions are obtained for the smooth operation of the CVs platoon with CAs, and the value of compensating coefficient corresponding to different attacks intensity are observed according to the Bode diagrams. Moreover, the traffic dynamics are investigated for different types of CAs including the decreased or magnified headway and velocity information, replay or delay acceleration and deceleration through numerical simulation. The simulation results reveal that the control term incorporating the effect of delayed headway and safety distance can successfully alleviate the negative influences of CAs on CVs platoon operations to guarantee traffic security and stability.
Internet of Medical Things (IoMT), an application of Internet of Things (IoT), is addressing countless limitation of traditional health-care systems such as quality of patient care, healthcare costs, ...shortage of medical staff and inadequate medical supplies in an efficient manner. With the use of the IoMT systems, there are unparalleled benefits that are enhancing the quality and efficiency of treatments and thereby are improving patients health. However, the 2018 Ransomware cyber-attack on Indiana hospital system exposed the critical fault-lines among IoMT environment. The gravity and frequency of cyber-attacks are expanding at an alarming rate. Motivated from aforementioned challenges, we propose an ensemble learning and fog-cloud architecture-driven cyber-attack detection framework for IoMT networks. The ensemble design, combines Decision Tree, Naive Bayes, and Random Forest as first-level individual learners. In the next level, the classification results are used by XGBoost for identifying normal and attack instances. Second, for dynamic and heterogeneous networks such as IoMT, fog, and cloud, we present a deployment architecture for the proposed framework as, Software as a Service (SaaS) in fog side and Infrastructure as a Service (IaaS) in cloud side. Further, most of the existing work is evaluated using KDD CUP99 or NSL-KDD dataset. These datasets lack modern IoMT-based attacks. Therefore, the proposed model uses a realistic dataset namely, ToN-IoT which is collected from a heterogeneous and large-scale IoT network. The experimental result shows that the proposed framework can achieve detection rate of 99.98%, accuracy of 96.35%, and can reduce false alarm rate up to 5.59%.
State estimation is one of the fundamental functions in modern power grid operations that provide operators with situational awareness and is used by several applications like contingency analysis ...and power markets. Several research in the recent past have highlighted the vulnerability of state estimators to stealthy false data injection attacks that bypass bad data detection mechanisms. They primarily focused on identifying stealthy attack vectors and characterizing their impacts on state estimates. Existing mitigation measures either focus on masking the effect of attacks through redundant measurements or prevent attacks by increasing the cyber security of associated sensors and communication channels. The solutions based on these offline approaches make specific assumptions about the nature of attacks and of the system, which are often restrictive and grossly inadequate to deal with dynamically evolving cyber threats and changing system configurations. In this paper, we propose an online anomaly detection algorithm that utilizes load forecasts, generation schedules, and synchrophasor data to detect measurement anomalies. We provide some insight into the factors that affect the performance of the proposed algorithm. We also describe an empirical method to obtain the minimum attack magnitudes and the detection thresholds for meeting specified false positive and true positive rates. Finally, we evaluated the performance of the proposed algorithm using the IEEE 14 bus power system model for several measures (false positive, false negative, and thresholds). We observed that the best performance of the proposed algorithm relies on finding the right balance between the minimum attack magnitude and detection thresholds. We also observed that the minimum attack magnitudes and detection thresholds could be further improved through the use of a combination of more accurate forecasts and PMU measurements.
