Attack Strategies Research Articles

Generation of security system defense strategies based on evolutionary game theory

The physical protection systems of Nuclear Power Plant are utilized to safeguard targets against intrude by attacker. As the methods employed by attackers to intrude Nuclear Power Plant become increasingly complex and diverse, there is an urgent need to identify optimal defense strategies to interrupt adversary intrusions. This paper focuses on studying the defense of security personnel against adversary intrusions and utilizes an evolutionary game approach to select the optimal defense decisions for physical protection systems. Under the assumption of bounded rationality for both the attacker and defender, the paper constructs replication dynamic equations for attack and defense strategies, investigating the process of strategy selection and the stability of evolution. Finally, a minimal model is proposed to validate the feasibility of utilizing the evolutionary game model for defense strategy selection.

State estimation for linear discrete-time systems with binary-valued quantized innovations against data tampering attacks

In today’s information era, cyber–physical systems (CPS) have been widely used in many industrial and infrastructure fields, and the study of security problems of CPS has become a crucial issue. For linear discrete-time systems, this paper investigates the state estimation problem based on the quantized innovations and transmission data tampering scenarios. From the attacker’s view, by designing the tampering matrix and combining the property of the error function, the optimal attack strategy in the sense of maximizing the prediction error covariance matrix is given. From the defender’s view, the optimal defense scheme based on multiplicative compensation is given in the case of a known attack strategy. Moreover, in the case of unknown attack strategies, a joint estimation algorithm is designed to give the optimal defense scheme subject to error constraints. In this paper, state estimators in recursive form are given for all above cases, and the algorithms are proved to be consistent. The theoretical results obtained are verified by simulations.

Efficient adversarial training with multi-fidelity optimization for robust neural network

Adversarial examples (AEs) pose a significant threat to the security and reliability of deep neural networks. Adversarial training (AT) is one of the effective defense methods, involving the integration of a number of generated AEs into the training process to enhance model robustness. However, the computational cost associated with AE generation is unbearable, particularly for large-scale tasks. In pursuit of fast AT, many algorithms generate AEs by adopting a simple attack strategy, but they often sacrifice the quality of AEs and suffer from catastrophic overfitting, resulting in suboptimal model robustness. To address these issues, our approach incorporates multi-fidelity optimization, which employs a dynamic attack strategy to generate AEs with varying fidelity within a suitable range. Furthermore, we introduce a surrogate-assisted fidelity estimation module at the beginning of our proposed algorithm, allowing for the adaptive determination of the fidelity range tailored to specific tasks. Comparative experiments with seven state-of-the-art algorithms on three networks and three datasets demonstrate that the proposed algorithm obtains a competitive robust accuracy but spends only 50% of the training time of the projected gradient descent algorithm.

Analysis of safe electricity consumption on load side based on attack and defence game model

AbstractWith the improvement of digitalization and intelligence in the power system, the safe operation of the power system is facing enormous challenges. The safe use of electricity on the load side is the key to achieving safe and reliable power system operation. The detection party needs amounts of human and material resources when the power network is attacked. In response to the current difficulties of low detection ability and high detection costs, this paper proposes an attack and defence game model that considers the differences between different nodes, ensuring the safety and economy of electricity consumption while reducing energy waste. At first, the structure of smart meters and the attack characteristics of intruders are summarized, and a basic attack and defence game model is constructed. The Nash equilibrium is then solved, and the optimal strategy for the game between the defender and the intruder is given to balance the relation between detection performance and energy consumption. In response to the differences generated by each node, strategies for attackers to launch attacks on different nodes and the setting of optimal thresholds for other nodes in the defence system are explored. Finally, case studies verify that the proposed model could reduce the cost of intruder detection while ensuring a specific detection rate.

Measurement-device-independent quantum random number generation over 23 Mbps with imperfect single-photon sources

Quantum randomness relies heavily on the accurate characterization of the generator implementation, where the device imperfection or inaccurate characterization can lead to incorrect entropy estimation and practical bias, significantly affecting the reliability of the generated randomness. Measurement-device-independent (MDI) quantum random number generation (QRNG) endeavors to produce certified randomness, utilizing uncharacterized and untrusted measurement devices that are vulnerable to numerous attack schemes targeting measurement loopholes. However, existing implementations have shown insufficient performance thus far. Here, we propose a high-speed MDI-QRNG scheme based on a robust measurement tomography approach against the imperfection of single-photon sources. Compared with the conventional approach, the decoy-state method is introduced to obtain more accurate tomography results and a tighter lower bound of randomness. Finally, by using a high-speed time-bin encoding system, we experimentally demonstrated the scheme and obtained a reliable min-entropy lower bound of 7.37×10−2 bits per pulse, corresponding to a generation rate over 23 Mbps, which substantially outperforms the existing realizations and makes a record in discrete-variable semi-device-independent QRNGs.

A multi-step attack path prediction method for oil & gas intelligence pipeline cyber physics system based on CPNE

Owing to the integration of the 5th Generation Mobile Communication Technology (5 G) and industrial internet construction, Oil & Gas intelligent pipeline network has witnessed accelerated construction and development. The Oil & Gas intelligent pipeline network proposed by China National Petroleum Corporation exhibits the advantages such as high integration, data unification, and intelligent decision-making, which helps to greatly reduce manual management costs, optimize risk decision-making by managers, and avoid risks such as human misuse. However, the highly integrated Cyber-physical system of the Internet and the Internet of Things faces various cyber security threats, and traditional intrusion detection systems exhibit drawbacks in the face of complex cyber-attack behaviors. In response to incomplete multi-step attack path prediction in traditional prediction methods, this paper proposes a multi-step attack path prediction method for Oil & Gas intelligence pipeline cyber physics system (CPS) based on Colored Petri nets and mixed strategy Nash equilibrium (CPNE). Taking into account the importance of assets and the dependency relationships between assets within the cyber-physical system of Oil & Gas intelligent pipeline network, CPNE constructs a node dependency model for the cyber-physical system of Oil & Gas intelligent pipeline network, simulates multi-step attack and defense games, calculates the probability of node failure triggering and transfer using mixed strategy Nash equilibrium, and extracts the risk of multi-step attack paths for different attack strategies. With the filtration and separation unit of a certain natural gas station as an example, this paper establishes a CPNE model to rank the risk of the first multi-step attack single attack source path and predict the risk of subsequent multi-step attack source paths, thus predicting multi-step attack paths. Taking three rounds of attacks as an example, compared to the paths predicted by traditional probability attack graph models, the proposed CPNE model is more advantageous in multi attack source path prediction, with a 2.33-fold increase in the relative coverage rate of path predictions.

Subsystem-importance-aware DoS attacks and countermeasures

This paper is concerned with the security issue of remote state estimation in multi-systems. Note that different subsystems or local state information possess varying levels of importance. A novel denial-of-service (DoS) attack type, referred to as a subsystem-importance-aware DoS attack, is proposed. The importance of a subsystem refers to the quantitative impact of its information on the correctness or accuracy of decision-making for the entire system. Here, the importance of a subsystem is positively correlated with the power allocated to transmit local information. Additionally, it is assumed that the attacker has access to the transmission power of all subsystems. The optimal attack strategies are then investigated under different subsystem-importance weight functions and the error functions that are related to the packet arrival rate, and the closed-form solutions under some special cases are provided. Furthermore, potential defense strategies are investigated, leveraging the gap in the interpretation of the subsystem importance between the attacker and the defender, formulated as a Stackelberg game. Finally, numerical examples are provided to illustrate the theoretical results and provide insights into possible countermeasures against our proposed attack.

Rumor Detection with A Novel Graph Neural Network Approach

The wide spread of rumors[3] on social media has caused a negative impact on people's daily life, leading to potential panic, fear and mental health problems for the public.[47] How to debunk rumors as early as possible remains a challenging problem. Existing studies mainly leverage information propagation structure to detect rumors[6], while very few works focus on correlation among users that they may coordinate to spread rumors in order to gain a large popularity. In this paper, we propose a new detection model, that jointly learns both the representations of user correlation and information propagation to detect rumors on social media. Specifically, we leverage graph neural networks to learn the representations of user correlation from a bipartite graph[5] that describes the correlations between users and source tweets[12], and the representations of information propagation with a tree structure. Then we combine the learned representations from these two modules to classify the rumors. Since malicious users intend to subvert our model after deployment, we further develop a greedy attack scheme to analyze the cost of three adversarial attacks: graph attack, comment attack and joint attack. Evaluation results on two public datasets illustrate that the proposed MODEL outperforms the state-of-the-art rumor detection models. We also demonstrate our method performs well for early rumor detection.[8] Moreover, the proposed detection method is more robust to adversarial attacks compared to the best existing method. Importantly, we show that it requires high cost for attackers to subvert user correlation pattern, demonstrating the importance of considering user correlation for rumor detection.

A Dual Stealthy Backdoor: From Both Spatial and Frequency Perspectives

Backdoor attacks pose serious security threats to deep neural networks (DNNs). Backdoored models make arbitrarily (targeted) incorrect predictions on inputs containing well-designed triggers, while behaving normally on clean inputs. Prior researches have explored the invisibility of backdoor triggers to enhance attack stealthiness. However, most of them only focus on the invisibility in the spatial domain, neglecting the generation of invisible triggers in the frequency domain. This limitation renders the generated poisoned images easily detectable by recent defense methods. To address this issue, we propose a DUal stealthy BAckdoor attack method named DUBA, which simultaneously considers the invisibility of triggers in both the spatial and frequency domains, to achieve desirable attack performance, while ensuring strong stealthiness. Specifically, we first use Wavelet Transform to embed the high-frequency information of the trigger image into the clean image to ensure attack effectiveness. Then, to attain strong stealthiness, we incorporate Fourier Transform and Cosine Transform to mix the poisoned image and clean image in the frequency domain. Moreover, DUBA adopts a novel attack strategy, training the model with weak triggers and attacking with strong triggers to further enhance attack performance and stealthiness. DUBA is evaluated extensively on four datasets against popular image classifiers, showing significant superiority over state-of-the-art backdoor attacks in attack success rate and stealthiness.

Boosting Adversarial Transferability across Model Genus by Deformation-Constrained Warping

Adversarial examples generated by a surrogate model typically exhibit limited transferability to unknown target systems. To address this problem, many transferability enhancement approaches (e.g., input transformation and model augmentation) have been proposed. However, they show poor performances in attacking systems having different model genera from the surrogate model. In this paper, we propose a novel and generic attacking strategy, called Deformation-Constrained Warping Attack (DeCoWA), that can be effectively applied to cross model genus attack. Specifically, DeCoWA firstly augments input examples via an elastic deformation, namely Deformation-Constrained Warping (DeCoW), to obtain rich local details of the augmented input. To avoid severe distortion of global semantics led by random deformation, DeCoW further constrains the strength and direction of the warping transformation by a novel adaptive control strategy. Extensive experiments demonstrate that the transferable examples crafted by our DeCoWA on CNN surrogates can significantly hinder the performance of Transformers (and vice versa) on various tasks, including image classification, video action recognition, and audio recognition. Code is made available at https://github.com/LinQinLiang/DeCoWA.

BadRL: Sparse Targeted Backdoor Attack against Reinforcement Learning

Backdoor attacks in reinforcement learning (RL) have previously employed intense attack strategies to ensure attack success. However, these methods suffer from high attack costs and increased detectability. In this work, we propose a novel approach, BadRL, which focuses on conducting highly sparse backdoor poisoning efforts during training and testing while maintaining successful attacks. Our algorithm, BadRL, strategically chooses state observations with high attack values to inject triggers during training and testing, thereby reducing the chances of detection. In contrast to the previous methods that utilize sample-agnostic trigger patterns, BadRL dynamically generates distinct trigger patterns based on targeted state observations, thereby enhancing its effectiveness. Theoretical analysis shows that the targeted backdoor attack is always viable and remains stealthy under specific assumptions. Empirical results on various classic RL tasks illustrate that BadRL can substantially degrade the performance of a victim agent with minimal poisoning efforts (0.003% of total training steps) during training and infrequent attacks during testing. Code is available at: https://github.com/7777777cc/code.

Mutual-Modality Adversarial Attack with Semantic Perturbation

Adversarial attacks constitute a notable threat to machine learning systems, given their potential to induce erroneous predictions and classifications. However, within real-world contexts, the essential specifics of the deployed model are frequently treated as a black box, consequently mitigating the vulnerability to such attacks. Thus, enhancing the transferability of the adversarial samples has become a crucial area of research, which heavily relies on selecting appropriate surrogate models. To address this challenge, we propose a novel approach that generates adversarial attacks in a mutual-modality optimization scheme. Our approach is accomplished by leveraging the pre-trained CLIP model. Firstly, we conduct a visual attack on the clean image that causes semantic perturbations on the aligned embedding space with the other textual modality. Then, we apply the corresponding defense on the textual modality by updating the prompts, which forces the re-matching on the perturbed embedding space. Finally, to enhance the attack transferability, we utilize the iterative training strategy on the visual attack and the textual defense, where the two processes optimize from each other. We evaluate our approach on several benchmark datasets and demonstrate that our mutual-modal attack strategy can effectively produce high-transferable attacks, which are stable regardless of the target networks. Our approach outperforms state-of-the-art attack methods and can be readily deployed as a plug-and-play solution.

FACL-Attack: Frequency-Aware Contrastive Learning for Transferable Adversarial Attacks

Deep neural networks are known to be vulnerable to security risks due to the inherent transferable nature of adversarial examples. Despite the success of recent generative model-based attacks demonstrating strong transferability, it still remains a challenge to design an efficient attack strategy in a real-world strict black-box setting, where both the target domain and model architectures are unknown. In this paper, we seek to explore a feature contrastive approach in the frequency domain to generate adversarial examples that are robust in both cross-domain and cross-model settings. With that goal in mind, we propose two modules that are only employed during the training phase: a Frequency-Aware Domain Randomization (FADR) module to randomize domain-variant low- and high-range frequency components and a Frequency-Augmented Contrastive Learning (FACL) module to effectively separate domain-invariant mid-frequency features of clean and perturbed image. We demonstrate strong transferability of our generated adversarial perturbations through extensive cross-domain and cross-model experiments, while keeping the inference time complexity.

SlowTrack: Increasing the Latency of Camera-Based Perception in Autonomous Driving Using Adversarial Examples

In Autonomous Driving (AD), real-time perception is a critical component responsible for detecting surrounding objects to ensure safe driving. While researchers have extensively explored the integrity of AD perception due to its safety and security implications, the aspect of availability (real-time performance) or latency has received limited attention. Existing works on latency-based attack have focused mainly on object detection, i.e., a component in camera-based AD perception, overlooking the entire camera-based AD perception, which hinders them to achieve effective system-level effects, such as vehicle crashes. In this paper, we propose SlowTrack, a novel framework for generating adversarial attacks to increase the execution time of camera-based AD perception. We propose a novel two-stage attack strategy along with the three new loss function designs. Our evaluation is conducted on four popular camera-based AD perception pipelines, and the results demonstrate that SlowTrack significantly outperforms existing latency-based attacks while maintaining comparable imperceptibility levels. Furthermore, we perform the evaluation on Baidu Apollo, an industry-grade full-stack AD system, and LGSVL, a production-grade AD simulator, with two scenarios to compare the system-level effects of SlowTrack and existing attacks. Our evaluation results show that the system-level effects can be significantly improved, i.e., the vehicle crash rate of SlowTrack is around 95% on average while existing works only have around 30%.

Grid Chaos: An uncertainty-conscious robust dynamic EV load-altering attack strategy on power grid stability

Electric Vehicles (EVs) and their charging infrastructure have witnessed rapid adoption into the smart grid owing to their ability to reduce the transportation sector's emissions. However, this hasty deployment has left the grid prone to attacks initiated through the charging ecosystem. This paper develops a family of robust dynamic load-altering attacks that manipulate the EV load to destabilize the power grid. The attack strategy is formulated based on feedback control theory and Linear Matrix Inequalities (LMIs) and incorporates mathematical modeling of the power grid uncertainties. The effectiveness of our proposed robust EV attack strategy is demonstrated through extensive simulations of attack scenarios against the New England (NE) 39-bus grid. The results demonstrate that the proposed attacks were able to destabilize the grid frequency inducing a deviation of 1.5 Hz and forcing the generators to trip after compromising as little as 6.6% of the total grid load. We also present a two-tier Deep Learning (DL) attack detection mechanism. We optimize this mechanism to achieve optimal results against dynamic and switching attacks. The proposed mechanism was evaluated against several other models and proved to be superior with an accuracy of 99.9%.

Infusing Morabaraba game design to develop a cybersecurity awareness game (CyberMoraba)

Numerous studies have confirmed the effectiveness of Cybersecurity Awareness Games (CAGs) in enhancing the security posture of diverse organizations. As these organizations increasingly face the formidable challenge of cyberattacks, implementing serious CAGs to solve this issue has become a paramount concern. This article introduces an innovative approach to cybersecurity education by presenting a serious CAG. The game aims to effectively educate students about critical aspects of cybersecurity awareness engagingly and interactively. The study aimed to redefine cybersecurity awareness training by introducing an indigenous game design that intricately incorporates the traditional South African Morabaraba board game. While the effectiveness of non-indigenous games like "Capture The Flag (CTF)" in cybersecurity training is acknowledged, indigenous designs have been overlooked. This research creatively integrates Morabaraba's gameplay into cybersecurity training, adapting it into a competitive game where players adopt the roles of either defenders or attackers, with corresponding tokens/images symbolizing various cyber defense and attack strategies. Both the defenders and attackers in the game can elevate their awareness scores by strategically positioning defensive or attacking images on the game board. Subsequently, a judging entity assesses the players' moves and assigns scores based on the accuracy of the images placed. The game mirrors real-world scenarios, promoting strategic thinking and leveraging interactive gameplay for practical insights into cybersecurity awareness. Players demonstrate their cybersecurity knowledge through offensive and defensive strategies. A group of 40 students evaluated the game's effectiveness, highlighting its potential to create an engaging and competitive learning environment that imparts cybersecurity principles and practical application. The evaluation of the game mechanics demonstrated a remarkably positive outcome, with students expressing both enjoyment and an enhanced understanding of cybersecurity awareness.

Robustness assessment of weakly coupled cyber-physical power systems under multi-stage attacks

Evaluating the robustness of cyber-physical system (CPS) under multi-stage attacks is fundamental to enhancing CPS resilience. Currently, many models inadequately consider information flow, which is not aligned with reality. The initial focus of this paper is on introducing a weakly coupled interdependent model for CPS. This model considers probabilistic line outage rules, cyber flow transmission characteristics, and the probabilities of inter-system coupling failures. Subsequently, the multi-stage attack strategies are presented and employed by the proposed model during disaster scenarios, along with the proposed robustness assessment indicators. The numerical results illustrate the robustness of the weakly-coupled CPS model under multi-stage attacks. This lays a modeling and simulation foundation for repair crew to enhance the CPS resilience.

Binary observation‐based identification for finite impulse response systems under denial of service attacks

SummaryThis article considers the parameter identification problem for the finite impulse response system based on binary observations under denial of service attacks. First, in addition to designing the identification algorithm for the unknown parameter, and its convergence is verified, simultaneously obtaining the asymptotic normality. Then, based on the convergence speed of the identification algorithm and the covariance matrix of the estimation error, the optimal attack strategy problem is transformed into an optimization problem with constraints, then the optimal solution is given. Furthermore, a defense strategy with dual time‐scale input design method is proposed and its effectiveness is demonstrated. Finally, numerical simulations are applied to evidence the correctness of the raised method.

Unpacking AI Security Considerations

The field of Artificial Intelligence has emerged as a convincing tool to be used in a myriad of applications like finance, traffic prediction, health and travel sectors. Due to the enormous benefits provided in terms of automation, convenience, processing time, reduced manhours, and productivity, AI is being seen as the next technical revolution. AI is being showcased as a useful tool to stimulate creativity as well as provide support with its tremendous computational power. The release of tools like ChatGPT has exploded onto the technological scene. Users are making use of Large Language Models (LLMs) and tools to perform a host of activities like writing an essay, translating documents, and finding travel plans. However, the popularity of these tools has not been without risk. In the technology marketplace, the race to dominance can force competitors to waive safety concerns in favour of product adoption. Many are unaware of the potential dangers and risks that may inherently reside within AI tools. This paper looks at the potential risks of AI tools such the creation of misinformation or scams. AI security has now become a paramount concern that should not be ignored. In this paper, the potential risks and threat vectors of Artificial Intelligence will be covered. The aim will be to provide insight into the malicious use of Artificial Intelligence Tools through a discussion of techniques to bypass security controls. The paper aims to provide a more detailed account on how AI can be manipulated in order to empower users about the latest attack schemes.

RoPE: Defending against backdoor attacks in federated learning systems

Federated learning (FL) is vulnerable to backdoor attacks, which aim to cause the misclassification on samples with a specific backdoor. Most existing algorithms are restricted by some conditions, such as the data distribution across the joint clients, the number of attackers and some auxiliary information, thereby being limited in the practical FL. In this paper, we propose RoPE containing three parts: using principal component analysis to extract the important features of model gradients; leveraging expectation–maximization to separate malicious clients from benign ones in accordance to the important features; removing the potential malicious gradients within the selected cluster with Isolated Forest. RoPE requires no restricted assumptions during the training process. We evaluate the performance of RoPE on three image classification tasks under non-independent and identically distributed scenario (non-iid) against centralized backdoor attacks with various ratios of attackers and distributed backdoor attacks, respectively. We also evaluate the performance of RoPE against other backdoor attack scenarios, including independent and identically distributed (iid) scheme, elaborately designed attack schemes. The results show that RoPE can defend against these backdoor attacks and outperform the existing algorithms. In addition, we also explore the impact of different numbers of features on RoPE’s performance and conduct ablation experiments.