Malicious Data Research Articles

Optimal attack strategy in system identification with binary‐valued observations: An information entropy‐based approach

AbstractWith the advancements in network technology and artificial intelligence, the interaction between network nodes is getting more frequent. The malicious data tampering can be extremely damaging to system stability, therefore the network security is of great importance. In this paper, we address the problem of system identification with binary‐valued observations under data tampering attacks and an information entropy‐based approach to optimal attack strategy from the viewpoint of attacker. Under the constrained attack energy, a method is proposed to obtain the optimal attack strategy using the information entropy as the optimization index. It is interestingly demonstrated that such optimal attack strategy is equivalent to the one by using the estimation error of the identification algorithm as the optimization index. The rationality of the proposed method is verified through numerical simulations.

Image sterilization via overwriting

Sterilization refers to the method of removing hidden information from images, effectively eradicating malicious data embedded within digital media. Existing methods are hampered by dual limitations: insufficient resilience to cutting-edge information hidden methods, coupled with the challenge of image restoration in quality. In response to these challenges, this paper proposes a novel method composed of overwriting attack and diffusion optimization stages. Employing the concept of pixel destruction, the overwriting method conducts a more thorough attack on hidden information while preserving sufficient information for image reconstruction and further diffusion optimization. This strategy is not bound to any particular information hiding method, making it effective against a wide range of steganography and watermarking techniques by systematically disrupting every pixel in the image. Diffusion optimization makes full use of the generative capabilities of diffusion models trained on large datasets, enabling higher quality image enhancements for a variety of scene images.Our method stands out in comparison to state-of-the-art studies, as experimental outcomes demonstrate its capability to maintain superior image integrity while efficiently eliminating hidden information across various steganography and watermarking techniques. Notably, when confronted with deep robust watermarking schemes, our approach exhibits a significant improvement over preceding methods. For instance, the removal rate improves significantly from 39.42% to 87.78% with higher quality restoration when compared to SC-Net. Our method delivers superior outcomes in the task of image sterilization, while presents a new challenge to the robustness of image information hiding methods and contributes to network security.

Robust Platoon Control of Mixed Autonomous and Human-Driven Vehicles for Obstacle Collision Avoidance: A Cooperative Sensing-Based Adaptive Model Predictive Control Approach

Obstacle detection and platoon control for mixed traffic flows, comprising human-driven vehicles (HDVs) and connected and autonomous vehicles (CAVs), face challenges from uncertain disturbances, such as sensor faults, inaccurate driver operations, and mismatched model errors. Furthermore, misleading sensing information or malicious attacks in vehicular wireless networks can jeopardize CAVs’ perception and platoon safety. In this paper, we develop a two-dimensional robust control method for a mixed platoon, including a single leading CAV and multiple following HDVs that incorporate robust information sensing and platoon control. To effectively detect and locate unknown obstacles ahead of the leading CAV, we propose a cooperative vehicle–infrastructure sensing scheme and integrate it with an adaptive model predictive control scheme for the leading CAV. This sensing scheme fuses information from multiple nodes while suppressing malicious data from attackers to enhance robustness and attack resilience in a distributed and adaptive manner. Additionally, we propose a distributed car-following control scheme with robustness to guarantee the following HDVs, considering uncertain disturbances. We also provide theoretical proof of the string stability under this control framework. Finally, extensive simulations are conducted to validate our approach. The simulation results demonstrate that our method can effectively filter out misleading sensing information from malicious attackers, significantly reduce the mean-square deviation in obstacle sensing, and approach the theoretical error lower bound. Moreover, the proposed control method successfully achieves obstacle avoidance for the mixed platoon while ensuring stability and robustness in the face of external attacks and uncertain disturbances.

Enhancing healthcare cloud security with intelligent shuffled frog leaping optimized sequential long short-term memory (ISFLO-SLSTM)

Although medical image processing consumes much memory, computing time and specialized platforms are needed. As a result, it is crucial to implement cost-effective solutions to replace outdated ones. Due to these factors, we choose to adopt cloud computing to satisfy our needs for large-scale data processing. Meanwhile, this strategy offers accessibility to services upon request with excellent reliability and versatility. Therefore, employing cloud services rather than internal applications will assist healthcare organizations in outsourcing calculations to a third party, thereby reducing operating costs. However, to avoid malicious data leaks, comprehensive protection of information across unauthorized users and untrustworthy clouds is necessary. Different frameworks are created to allow consumers to save and analyze their information utilizing cloud-based computing. They are developed utilizing distributed systems, cryptosystems, or an amalgam of the two. The primary issue with applying cryptography methods for the enormous analysis of information on the cloud is the computing cost of image processing operations. Preventing unauthorized usage of healthcare records and private health details is the main concern. To secure data processing in a cloud environment by classifying image pixels, we offer a revolutionary intelligent shuffled frog jumping optimized sequential long short-term memory (ISFLO-SLSTM) approach. We added an additional layer, the cloud security module, to lower the danger of future medical data leakage. The effectiveness of the suggested strategy is assessed using medical RGB images taken from the Kaggle database. The findings of this study show how the ISFLO-SLSTM system has the potential to revolutionize healthcare information analysis by protecting patient privacy and security that utilizes the full extent of the latest machine learning methods.

Classification of DDoS attack traffic on SDN network environment using deep learning

Distributed Denial of Service (DDoS) attack is a major threat to the Internet of Things (IoT), Software Defined Networks (SDN), and Cloud Computing Networks. Due to the tremendous applications of IoT networks, the number of DDoS attacks is increasing significantly, and most sophisticated DDoS attacks are generated through IoT botnets. An IoT botnet-based DDoS attack can disrupt the network quickly with a surge of malicious traffic. Especially in an SDN network, it is important to detect the DDoS attack before it occurs to the SDN controller. DDoS attacks on the centralized controller of the SDN can disrupt the whole network. So, identifying DDoS attacks at the earliest is a critical security measure for network experts and practitioners. In this paper, we analyze the DDoS attack on an SDN environment and develop a method to identify the DDoS attack using Deep Learning (DL). The proposed method converts the captured raw network traffic to image data and classifies the malicious data from normal data. The method is evaluated on our test-bed simulated dataset and two other benchmark datasets. The experimental comparison shows that the proposed method performs better on all three datasets, giving more than 99% classification accuracy.

Optimal strategy of data tampering attacks for FIR system identification with average entropy and binary‐valued observations

SummaryIn the era of digitalization boom, cyber‐physical system (CPS) has been widely used in several fields. However, malicious data tampering in communication networks may lead to degradation of the state estimation performance, which may affect the control decision and cause significant losses. In this paper, for the identification of finite impluse response (FIR) systems with binary‐valued observations under data tampering attack, an optimal attack strategy based on the average entropy is designed from the perspective of the attacker. In the case of unknown parameters, the regression matrix is used to give the estimation method of the system parameters, the algorithmic flow of the data tampering attack for the implementation of the on‐line attack is designed. Finally, the effectiveness of the algorithm and the reliability of the conclusions is verified through the examples.

Efficient Post-Quantum Pattern Matching on Encrypted Data

Pattern matching methods are essential in various applications where users must disclose highly sensitive information. Among these applications are genomic data analysis, financial records inspection, and intrusion detection processes, all of which necessitate robust privacy protection mechanisms. Balancing the imperative of protecting the confidentiality of analyzed data with the need for efficient pattern matching presents a significant challenge. In this paper, we propose an efficient post-quantum secure construction that enables arbitrary pattern matching over encrypted data while ensuring the confidentiality of the data to be analyzed. In addition, we address scenarios where a malicious data sender, intended to send an encrypted content for pattern detection analysis, has the ability to modify the encrypted content. We adapt the data fragmentation technique to handle such a malicious sender. Our construction makes use of a well-suited Homomorphic Encryption packing method in the context of fragmented streams and combines homomorphic operations in a leveled mode (i.e. without bootstrapping) to obtain a very efficient pattern matching detection process. In contrast to the most efficient state-of-the-art scheme, our construction achieves a significant reduction in the time required for encryption, decryption, and pattern matching on encrypted data. Specifically, our approach decreases the time by factors of 1850 , 10 6 , and 245 , respectively, for matching a single pattern, and by factors of 115 , 10 5 , and 12 , respectively, for matching 2 10 patterns.

Machine learning-based detection of the man-in-the-middle attack in the physical layer of 5G networks

Fifth generation communication networks (5G) has received a great deal of attention from academia and industry alike, which will enable a wide variety of vertical applications by connecting heterogeneous devices and machines. Assessing availability and reliability in many circumstances and environments is critical. Researchers have recently focused on investigating and analyzing new multimedia networks with artificial intelligence (AI) technologies to achieve higher data rates and secure communication traffic between parties. User information privacy and security are of vital importance and of growing concerns that present evolving challenges to overcome in preventing attacks. Man-in-the-middle (MITM) attack is considered one of the most common attacks, where an attacker can impersonate one of the parties in a communication system to steal user data or forge the malicious data. Due to the limitation of using conventional cryptographic techniques for mobile networks and similar systems, new methods have been introduced to validate and authenticate transmitted signals dynamically, depending on the physical layer. In this paper, we present the distance-time directional delay (DTDD) model to detect the MITM attack in a variety of contexts and scenario. Indoor hotspots (InH) and urban micro-cellular (UMi) propagation environments were investigated to verify the reliability of the proposed approaches using realistic 5G millimeter-wave configurations and system setups. Simulations have been constructed based on the mmWave 5G channel simulator tool NYUSIM, in conjunction with a collection of machine learning algorithms (ML) including the extreme gradient boosting (XGBoost) and light gradient boosting machine (LGBM) as the core of the presented models and methodologies. Numerical simulations results produced a detection accuracy approaching 100% in the InH environment scenario, whereas for UMi environment scenario, a detection accuracy approaching 99% was attained.

Enhancing security in smart healthcare systems: Using intelligent edge computing with a novel Salp Swarm Optimization and radial basis neural network algorithm

A smart healthcare system (SHS) is a health service system that employs advanced technologies such as wearable devices, the Internet of Things (IoT), and mobile internet to dynamically access information and connect people and institutions related to healthcare, thereby actively managing and responding to medical ecosystem needs. Edge computing (EC) plays a significant role in SHS as it enables real-time data processing and analysis at the data source, which reduces latency and improves medical intervention speed. However, the integration of patient information, including electronic health records (EHRs), into the SHS framework induces security and privacy concerns. To address these issues, an intelligent EC framework was proposed in this study. The objective of this study is to accurately identify security threats and ensure secure data transmission in the SHS environment. The proposed EC framework leverages the effectiveness of Salp Swarm Optimization and Radial Basis Functional Neural Network (SS-RBFN) for enhancing security and data privacy. The proposed methodology commences with the collection of healthcare information, which is then pre-processed to ensure the consistency and quality of the database for further analysis. Subsequently, the SS-RBFN algorithm was trained using the pre-processed database to distinguish between normal and malicious data streams accurately, offering continuous monitoring in the SHS environment. Additionally, a Rivest-Shamir-Adelman (RSA) approach was applied to safeguard data against security threats during transmission to cloud storage. The proposed model was trained and validated using the IoT-based healthcare database available at Kaggle, and the experimental results demonstrated that it achieved 99.87 % accuracy, 99.76 % precision, 99.49 % f-measure, 98.99 % recall, 97.37 % throughput, and 1.2s latency. Furthermore, the results achieved by the proposed model were compared with the existing models to validate its effectiveness in enhancing security.

An Efficient Lightweight Crypto Security Module for Protecting Data Transmission Through IOT Based Electronic Sensors

The Internet of Things (IoT) devices are advanced nanoelectronics devices which has recently witnessed an explosive expansion in the field of communication and electronics, becoming ubiquitous in various applications. However, the rapid growth of IoT applications makes them prone to security threats and data breaches. Hence, cryptographic techniques are developed to ensure data confidentiality and integrity in IoT and many of the applications from optoelectronics. However, the existing cryptographic algorithms face challenges in securing the data from threats during transmission, as they lack effective key management. Therefore, we proposed a novel optimized lightweight cryptography (LWC) to resolve this challenge using the combined benefits of Grey Wolf Optimization and Hyper Elliptic Curve Cryptography (GW-HECC). The proposed LWC algorithm protects the data from attacks during data exchange by optimizing the key management process and aims to deliver greater Quality of Service (QoS) in IoT networks. An IoT network was initially created with multiple sensor devices, IoT gateways, and data aggregators. The proposed framework includes a Quantum Neural Network (QNN)-based attack prediction module to predict the malicious data entry in the IoT network. The QNN learns the attack patterns from the historical IoT data and prevents incoming malicious data entries, ensuring that only normal data is transmitted to the cloud. For secure data transmission, the sensed data from the IoT network are encrypted using the proposed GW-HECC. The presented work was designed and implemented in Python software; the experimental results demonstrate that the proposed method offers greater data confidentiality of 97.9%, improved attack prediction accuracy of 99.8%, and a reduced delay of 0.37 s. Furthermore, a comparative analysis was made with existing cryptographic algorithms, manifesting that the proposed algorithm acquired improved results.

Enhancing Cyber Threat Detection with an Improved Artificial Neural Network Model

Identifying cyber-attacks that attempt to compromise digital systems is a critical function of intrusion detection systems (IDS). Data labeling difficulties, incorrect conclusions, and vulnerability to malicious data injections are only a few drawbacks of using machine learning algorithms for cybersecurity. To overcome these obstacles, researchers have created several network IDS models, such as the Hidden Naive Bayes Multiclass Classifier and supervised/unsupervised machine learning techniques. This study provides an updated learning strategy for artificial neural network (ANN) to address data categorization problems caused by unbalanced data. Compared to traditional approaches, the augmented ANN’s 92% accuracy is a significant improvement owing to the network’s increased resilience to disturbances and computational complexity, brought about by the addition of a random weight and standard scaler. Considering the ever-evolving nature of cybersecurity threats, this study introduces a revolutionary intrusion detection method.

Spatiotemporal polynomial graph neural network for anomaly detection of complex systems

Internet of Things systems are developed rapidly, there is a need for anomaly detection to identify malicious data to guarantee that the systems perform under healthy condition. However, the dynamic system state is not only influenced by sequential patterns in the time dimension but also affected by other sensor channels in the spatial dimension. Although many successful models have been established in past research to handle multivariate time series data, most of these models exhibit shortcomings in capturing spatiotemporal dependencies. The Internet of Things offers a promising platform for advanced data analysis, but its unique characteristics present challenges for anomaly detection. This paper introduces a novel spatiotemporal polynomial graph neural network (STPGNN) is proposed for anomaly detection of complex systems within the Internet of Things (IoT) framework. Specifically, we propose an adaptive spatiotemporal feature modeling module that includes two developed adaptive modules: the Adaptive Temporal Module and the Spatial Module. These modules are designed to capture dynamic spatial dependencies across multiple time steps and temporal dependencies over time, respectively. The results are then output through a decoder. Importantly, the multi-head attention mechanism employed in both sub-modules can effectively explore potential spatiotemporal dependency patterns within different subspaces. The proposed model, utilizing the innovative approach of STPGNN, demonstrates superior performance compared to existing techniques, as evidenced by experimental results on the real-world test run dataset. This marks a significant advancement in the field of anomaly detection, particularly for complex systems.

Computer network big data detection based on internet of things technology

In order to solve the problems of traditional network security intrusion detection system in detection accuracy and missed detection rate, this paper proposes a computer network data detection based on Internet of things technology. This paper designs an intrusion detection system based on data clustering analysis algorithm. The hardware part of the intrusion detection system is composed of six parts: detection module, adaptive module, control management module, risk early warning module, control access module and data acquisition module. Each module is closely connected and updated with the rule base and database in real time; In the software flow design of the system, based on the data clustering analysis algorithm, the collected original data are classified and processed, and the Mings distance is determined, revealing the difference between normal data and abnormal data, and finally realizing the accurate detection of cloud computing network intrusion behavior. The experimental results show that the proposed intrusion detection system has a higher detection accuracy for malicious data, and the average missed detection rate can be controlled below 0.3 %. Conclusionthe detection system designed in this paper has more advantages in stability performance.

Edge server enhanced secure and privacy preserving federated learning

Federated learning (FL) has been smoothly embedded into current IoT edge computing architecture, and hatching advanced IoT-FL applications. While so, sensitive messages generated on terminal sides brings privacy issues; and vulnerable terminal devices are easy targets for Byzantine adversaries, they inject malicious data to manipulate the IoT-FL system. Now an inherent dilemma is, the privacy preserving requires anonymous indistinguishable individual characteristics, while the malicious adversary detection requires transparent distinguishable results. Existing fragmented schemes are not able to handle both problems coherently. This time, inherit our previous study, based on the IoT edge computing architecture, we utilize Homomorphic Encryption (HE) and Threshold Secret Sharing (SS) methods, propose a joint scheme (namely, Sec-IoTFL) for anonymous adversary detection. Specifically, we batch encode the clients’ training result vectors (message) as polynomials, and split them as secret pieces with a unified SS key set; then edge server and cloud server joint deal with these secret pieces in a specialized flow, where Byzantine adversaries will be filtered out. The theoretical analysis and solid experimental results suggest, in our scheme, local sensitive data is well preserved, and malicious behavior is precisely screened. Comparing with traditional methods, our Sec-IoTFL scheme shows the superiority in accuracy and efficiency.

Hybrid Cyber-Physical Intrusion Detection System for Smart Manufacturing

Smart manufacturing is an important part of our critical infrastructure and is the current age of industry where physical components such as robotic arms, 3-D Printers, CNC machine, etc. are all interconnected and remotely controlled or automated to provide a major boost in efficiency. While being more effective, the cyber-physical integration expands the attack surface of these systems for any potential threats to act on and exploit. This integration also creates gaps in the current intrusion detection systems (IDS) and the research of such systems as they focus on either the cyber or physical components of these system, which leaves blind spots when an attack can only be detected by using either cyber or physical features. This paper fill that research gap by creating a cyber-physical testbed, launching denial of service and physical hijacking attacks, collecting benign and malicious data, and creating a hybrid IDS using K-Nearest Neighbors and Decision Tree models that consider both cyber and physical features. Our proposed hybrid IDS achieves an accuracy of 97.2% which was roughly the same as separate cyber and physical IDSs, but there was a significant boost in precision (98.4%), recall (94.2%), and F1 score (96.1%) when using the hybrid IDS compared to the separate IDSs.

TwI-FTM: Two-way IoT-FoG trust management scheme for task offloading in IoT-FoG networks

Based on trust management, the service establishment between IoT/User devices and fog networks reduces the impact of malicious data and nodes. Hence, building a trusted environment in the IoT and fog layers is very important. There is no assurance that the generated data or task is standard data; it may also have abnormal data to modify its originality. Similarly, the nodes in the fog environment also have malicious nodes during task offloading. For this, the novel Two-way IoT data & Fog Trust management (TwI-FTM) algorithm is proposed to identify the abnormal data with the help of the Trusted Data Measurement (TDM) and Fog Access Trust Evaluation (FATE) method for identifying the malicious fog access nodes in fog network. Identifying these abnormal data and malicious fog access nodes is done by continuously monitoring fog control nodes with the threshold level. Once it is identified, they will automatically be removed from the IoT-FoG network. The obtained trust measurements are analyzed and compared with other existing methods; it is identified that the proposed TwI-FTM outperforms and has a higher trust degree of 0.68 compared with others. It achieved higher detection accuracy of malicious fog access nodes in fog networks at 21 %, 16 %, and 14 % compared with OTM, TTM, and SLA-Trust, respectively. Finally, the task completion time was also reduced, and it was achieved at 4.5 ms, 5 ms, and 13 ms compared with TTM, OTM, and SLA-Trust, respectively.

An Artificial Intelligence-Based Intrusion Detection System using Optimization and Deep Learning

The Intrusion Detection System (IDS) was developed to resolve the malicious activities on the networking. However, the existing IDS models face issues in achieving high accuracy, and it is prone to detection error. Hence, to address these challenges, an innovative Vulture-based Deep Belief System (VbDBNS) for enhancing the performance of detecting intrusion activity by monitoring their behaviour and features. This study utilized the NSL-KDD database for training and validating the system. Henceforth, pre-processing is employed to standardize the dataset using Min-Max normalization, and 1-N encoding, which transforms the categorical data variables. Moreover, feature extraction was utilized to capture the most significant 42 attributes from the pre-processed database. Further, a classification module was designed using VbDBN to categorize normal and malicious data in which the fitness of vulture was updated, which enables it to monitor and detect the intrusion continuously. The experimental results of the study illustrate that the designed methodology achieved better outcomes than the conventional techniques in terms of recall, accuracy, precision, F1-score, etc.

An Adaptive Temporal Convolutional Network Autoencoder for Malicious Data Detection in Mobile Crowd Sensing.

Mobile crowdsensing (MCS) systems rely on the collective contribution of sensor data from numerous mobile devices carried by participants. However, the open and participatory nature of MCS renders these systems vulnerable to adversarial attacks or data poisoning attempts where threat actors can inject malicious data into the system. There is a need for a detection system that mitigates malicious sensor data to maintain the integrity and reliability of the collected information. This paper addresses this issue by proposing an adaptive and robust model for detecting malicious data in MCS scenarios involving sensor data from mobile devices. The proposed model incorporates an adaptive learning mechanism that enables the TCN-based model to continually evolve and adapt to new patterns, enhancing its capability to detect novel malicious data as threats evolve. We also present a comprehensive evaluation of the proposed model's performance using the SherLock datasets, demonstrating its effectiveness in accurately detecting malicious sensor data and mitigating potential threats to the integrity of MCS systems. Comparative analysis with existing models highlights the performance of the proposed TCN-based model in terms of detection accuracy, with an accuracy score of 98%. Through these contributions, the paper aims to advance the state of the art in ensuring the trustworthiness and security of MCS systems, paving the way for the development of more reliable and robust crowdsensing applications.

Investigating and mitigating blockchain poisoning attacks

With the burgeoning adoption of smart contracts and the intrinsic anonymity that blockchain technology provides, the past few years have witnessed an exponential surge in the deployment of blockchain platforms for managing crypto assets. While Ether stands as a paragon of blockchains equipped with smart contract capabilities, its noteworthy to mention that such blockchains present a blank canvas in the form of a data list, dedicated for unbridled program storage. This vacant space, though pivotal for various operations, can become a potential Achilles heel. Specifically, if this repository gets tainted with malicious data, the repercussions can reverberate across the entire blockchain, disrupting its core functions. The research in focus shines a spotlight on a particularly insidious threat, termed blockchain poisoning attack. Through this stratagem, malevolent actors can, with minimal expense, embed detrimental files within the data space, thereby extensively polluting the blockchain. This isnt just a theoretical hazard; its practical implications are stark and can jeopardize the integrity of countless transactions. In a commendable feat of investigative rigor, the research meticulously decoded the modus operandi behind these blockchain poisoning attacks. By employing a multi-pronged research approach, encompassing in-depth studies and iterative inquiries, it successfully deconstructed the exact techniques that malefactors employ.

Enhancing Privacy and Security in Large-Language Models: A Zero-Knowledge Proof Approach

The explosive growth of Large-Language Models (LLMs), particularly Generative Pre-trained Transformer (GPT) models, has revolutionised fields ranging from natural language processing to creative writing. Yet, their reliance on vast, often unverified data sources introduces a critical vulnerability: unreliability and security concerns. Traditional GPT models, while impressive in their capabilities, struggle with limited factual accuracy and susceptibility to manipulation by biased or malicious data. This poses a significant risk in professional and personal environments where sensitive or mission-critical data is paramount. This work tackles this challenge head-on by proposing a novel approach to enhance GPT security and reliability: leveraging Zero-Knowledge Proofs (ZKPs). Unlike traditional cryptographic methods that require sensitive data exchange, ZKPs allow one party to convincingly prove the truth of a statement, without revealing the underlying information. In the context of GPTs, ZKPs can validate the legitimacy and quality of data sources used in GPT computations, combating data manipulation and misinformation. This ensures trustworthy outputs, even when incorporating third-party data (TPD). ZKPs can securely verify user identities and access privileges, preventing unauthorised access to sensitive data and functionality. This protects critical information and promotes responsible LLM usage. ZKPs can identify and filter out manipulative prompts designed to elicit harmful or biased responses from GPTs. This safeguards against malicious actors and promotes ethical LLM development. ZKPs facilitate training specialised GPT models on targeted datasets, resulting in deeper understanding and more accurate outputs within specific domains. This allows the creation of ‘expert-GPT’ applications in specialised fields like healthcare, finance, and legal services. The integration of ZKPs into GPT models represents a crucial step towards overcoming trust and security barriers. Our research demonstrates the viability and efficacy of this approach, with our ZKP-based authentication system achieving promising results in data verification, user control, and malicious prompt detection. These findings lay the groundwork for a future where GPTs, empowered by ZKPs, operate with unwavering integrity, fostering trust and accelerating ethical AI development across diverse domains.