Malicious Data Research Articles

Veritas: Layer-2 Scaling Solution for Decentralized Oracles on Ethereum Blockchain with Reputation and Real-Time Considerations

Blockchainsand smart contracts are pivotal in transforming interactions between systems and individuals, offering secure, immutable, and transparent trust-building mechanisms without central oversight. However, Smart Contracts face limitations due to their reliance on blockchain-contained data, a gap addressed by ’Oracles’. These bridges to external data sources introduce the ’Oracle problem’, where maintaining blockchain-like security and transparency becomes vital to prevent data integrity issues. This paper presents Veritas, a novel decentralized oracle system leveraging a layer-2 scaling solution, enhancing smart contracts’ efficiency and security on Ethereum blockchains. The proposed architecture, explored through simulation and experimental analyses, significantly reduces operational costs while maintaining robust security protocols. An innovative node selection process is also introduced to minimize the risk of malicious data entry, thereby reinforcing network security. Veritas offers a solution to the Oracle problem by aligning with blockchain principles of security and transparency, and demonstrates advancements in reducing operational costs and bolstering network integrity. While the study provides a promising direction, it also highlights potential areas for further exploration in blockchain technology and oracle system optimization.

A Framework for Detecting False Data Injection Attacks in Large-Scale Wireless Sensor Networks.

False data injection attacks (FDIAs) on sensor networks involve injecting deceptive or malicious data into the sensor readings that cause decision-makers to make incorrect decisions, leading to serious consequences. With the ever-increasing volume of data in large-scale sensor networks, detecting FDIAs in large-scale sensor networks becomes more challenging. In this paper, we propose a framework for the distributed detection of FDIAs in large-scale sensor networks. By extracting the spatiotemporal correlation information from sensor data, the large-scale sensors are categorized into multiple correlation groups. Within each correlation group, an autoregressive integrated moving average (ARIMA) is built to learn the temporal correlation of cross-correlation, and a consistency criterion is established to identify abnormal sensor nodes. The effectiveness of the proposed detection framework is validated based on a real dataset from the U.S. smart grid and simulated under both the simple FDIA and the stealthy FDIA strategies.

A Sanitizable Access Control With Policy-Protection for Vehicular Social Networks

As an emerging field of communication, Vehicular Social Networks (VSNs) can reduce traffic congestion while enhancing road safety by sharing data among groups of commuters. In VSNs, Vehicular Cloud Server (VCS) based data sharing technology with encrypted primitives allows local users to outsource encrypted data for reducing the storage burden on the user side and sharing data without location restrictions. However, existing data encryption solutions that have been applied in VSNs environments still encounter weaknesses in efficiency, security, or privacy due to the following problems: (1) lack of effective access policies for flexible authorizing ciphertext to multiple data users; (2) data breaches caused by malicious data publishers; (3) necessity in hiding the private information of receivers. To date, no such solution has been available that securely enables one-to-many user authorization with privacy protection, while greatly resisting malicious data publishers. We propose a Sanitizable Access Control System with Policy-protection (SASP) for VSNs in this paper. Our SASP enables a sanitizer to test and sanitize encrypted data to defend against malicious data publishers, ensuring that the plaintext can only be recovered if an authorized user has a valid key. Furthermore, in our SASP system, the access policy is separated into attribute names and attribute values. Wherein, the attribute values contain a lot of private information, which is hidden in the ciphertext to guarantee data users’ privacy. Rigorous security analysis and performance evaluations demonstrate the practicality of SASP for VSNs.

An Efficient Privacy Protection Mechanism for Blockchain-Based Federated Learning System in UAV-MEC Networks.

The widespread use of UAVs in smart cities for tasks like traffic monitoring and environmental data collection creates significant privacy and security concerns due to the transmission of sensitive data. Traditional UAV-MEC systems with centralized data processing expose this data to risks like breaches and manipulation, potentially hindering the adoption of these valuable technologies. To address this critical challenge, we propose UBFL, a novel privacy-preserving federated learning mechanism that integrates blockchain technology for secure and efficient data sharing. Unlike traditional methods relying on differential privacy (DP), UBFL employs an adaptive nonlinear encryption function to safeguard the privacy of UAV model updates while maintaining data integrity and accuracy. This innovative approach enables rapid convergence, allowing the base station to efficiently identify and filter out severely compromised UAVs attempting to inject malicious data. Additionally, UBFL incorporates the Random Cut Forest (RCF) anomaly detection algorithm to actively identify and mitigate poisoning data attacks. Extensive comparative experiments on benchmark datasets CIFAR10 and Mnist demonstrably showcase UBFL's effectiveness. Compared to DP-based methods, UBFL achieves accuracy (99.98%), precision (99.93%), recall (99.92%), and F-Score (99.92%) in privacy preservation while maintaining superior accuracy. Notably, under data pollution scenarios with varying attack sample rates (10%, 20%, and 30%), UBFL exhibits exceptional resilience, highlighting its robust capabilities in securing UAV gradients within MEC environments.

Novel nature-inspired optimization approach-based svm for identifying the android malicious data

Novel nature-inspired optimization approach-based svm for identifying the android malicious data

A Hybrid Cryptographic Mechanism for Secure Data Transmission in Edge AI Networks

As Edge AI systems become more prevalent, ensuring data privacy and security in these decentralized networks is essential. In this work, a novel hybrid cryptographic mechanism was presented by combining Ant Lion Optimization (ALO) and Diffie–Hellman-based Twofish cryptography (DHT) for secure data transmission. The developed work collects the data from the created edge AI system and processes it using the Autoencoder. The Autoencoder learns the data patterns and identifies the malicious data entry. The Diffie–Hellman (DH) key exchange generates a shared secret key for encryption, while the ALO optimizes the key exchange and improves security performance. Further, the Twofish algorithm performs the data encryption using a generated secret key, preventing security threats during transmission. The implementation results of the study show that it achieved a higher accuracy of 99.45%, lower time consumption of 2 s, minimum delay of 0.8 s, and reduced energy consumption of 3.2 mJ.

The Threat of Disruptive Jamming to Blockchain-Based Decentralized Federated Learning in Wireless Networks.

Machine learning techniques have attracted considerable attention for wireless networks because of their impressive performance in complicated scenarios and usefulness in various applications. However, training with and sharing raw data obtained locally from each wireless node does not guarantee privacy and requires a large communication overhead. To mitigate such issues, federated learning (FL), in which sharing parameters for model updates are shared instead of raw data, has been developed. FL has also been studied using blockchain techniques to efficiently perform learning in distributed wireless systems without having to deploy a centralized server. Although blockchain-based decentralized federated learning (BDFL) is a promising technique for various wireless sensor networks, malicious attacks can still occur, which result in performance degradation or malfunction. In this study, we analyze the impact of a jamming threats from malicious miners to BDFL in wireless networks. In a wireless BDFL system, it is possible for malicious miners with jamming capability to interfere with the collection of model parameters by normal miners, thus preventing the victim miner from generating a global model. By disrupting normal miners participating in BDFL systems, malicious miners with jamming capability can more easily add malicious data to the mainstream. Through various simulations, we evaluated the success probability performance of malicious block insertion and the participation rate of normal miners in a wireless BDFL system.

Observer‐based finite‐time security fault‐tolerant control for nonlinear multi‐agent systems under byzantine adversaries

AbstractThis article investigates the issue of security‐based adaptive output feedback finite‐time fault‐tolerant control (FTC) for nonlinear multi‐agent systems subjected to Byzantine attacks. In this work, the fuzzy logic systems are employed to approximate the unknown nonlinearities, and unmeasurable states are estimated by the designed state observer. In addition, a data selector is designed to detect extremely malicious data of the controlled system. Considering the actuator faults, a novel Nussbaum function is utilized to compensate for the uncertainties of attacks and intermittent failures. Then, by introducing first‐order filter, an observer‐based distributed adaptive output feedback finite‐time security FTC algorithm is developed, which demonstrates all signals of the system are bounded, and the feasibility and effectiveness of the developed control scheme can be verified by simulation results.

A Sampling-Based Method for Detecting Data Poisoning Attacks in Recommendation Systems

The recommendation algorithm based on collaborative filtering is vulnerable to data poisoning attacks, wherein attackers can manipulate system output by injecting a large volume of fake rating data. To address this issue, it is essential to investigate methods for detecting systematically injected poisoning data within the rating matrix. Since attackers often inject a significant quantity of poisoning data in a short period to achieve their desired impact, these data may exhibit spatial proximity. In other words, poisoning data may be concentrated in adjacent rows of the rating matrix. This paper capitalizes on the proximity characteristics of poisoning data in the rating matrix and introduces a sampling-based method for detecting data poisoning attacks. First, we designed a rating matrix sampling method specifically for detecting poisoning data. By sampling differences obtained from the original rating matrix, it is possible to infer the presence of poisoning attacks and effectively discard poisoning data. Second, we developed a method for pinpointing malicious data based on the distance of rating vectors. Through distance calculations, we can accurately identify the positions of malicious data. After that, we validated the method on three real-world datasets. The results demonstrate the effectiveness of our method in identifying malicious data within the rating matrix.

Backdozer : A Backdoor Detection Methodology for DRL-based Traffic Controllers

While the advent of Deep Reinforcement Learning (DRL) has substantially improved the efficiency of Autonomous Vehicles (AVs), it makes them vulnerable to backdoor attacks that can potentially cause traffic congestion or even collisions. Backdoor functionality is typically implanted by poisoning training datasets with stealthy malicious data, designed to preserve high accuracy on legitimate inputs while inducing desired misclassifications for specific adversary-selected inputs. Existing countermeasures against backdoors predominantly concentrate on image classification, utilizing image-based properties, rendering these methods inapplicable to the regression tasks of DRL-based AV controllers that rely on continuous sensor data as inputs. In this paper, we introduce the first-ever defense against backdoors on regression tasks of DRL-based models, called Backdozer . Our method systematically extracts more abstract features from representations of training data by projecting them into a specific latent subspace and segregating them into several disjoint groups based on the distribution of legitimate outputs. The key observation of Backdozer is that authentic representations for each group reside in one latent subspace, whereas the incorporation of malicious data impacts that subspace. Backdozer optimizes a sample-wise weight vector for the representations capturing the disparities in projections originating from different groups. We experimentally demonstrate that Backdozer can attain \(100\% \) accuracy in detecting backdoors. We also evaluate its effectiveness against three closely related state-of-the-art defenses.

Quantum Enhanced Hazardous Substances Surveillance System

Unauthorized access to hazardous substances poses a serious threat to public health. Typically, these substances, such as chemicals, are housed within warehouses or laboratory facilities, making strict control over any unauthorized entry. The control can be achieved through various measures, including access control systems, registration systems, remote inspection and surveillance, and personnel checks. Given its essential significance, the security of surveillance system data transmission plays a critical role. On the one hand, authorized entities must be capable of detecting any malicious data interception, and on the other hand, confidential information must remain safeguarded when an eavesdropper is identified. To address transmission security, we have developed a quantum-communication-based system for transmitting surveillance imagery of the hazardous chemical storage area entrance. This system effectively thwarts eavesdropping and data tampering during transmission, thereby enhancing the security of conventional monitoring systems.

Edge-based detection and localization of adversarial oscillatory load attacks orchestrated by compromised EV charging stations

Recent reports indicate that electric vehicle charging stations (EVCSs) are susceptible to remote exploitation through their vulnerable software/cyber components. More importantly, compromised EVCS can be leveraged to perform coordinated oscillatory load attacks against the interconnected power grid, leading to power grid instability, increased operational costs, and power line tripping. In this paper, an edge-based approach for the detection and localization of coordinated oscillatory load attacks initiated by exploited EV charging stations against the power grid is investigated. The edge-based detection relies on the behavioral characteristics of the power grid in the presence of interconnected EVCS while combining cyber and physical layer features to implement deep learning algorithms for the effective detection of oscillatory load attacks at the EVCS. The proposed detection approach was evaluated by building a real-time test bed to synthesize benign and malicious data, which was generated by analyzing real-life EV charging data collected during recent years. The results demonstrate the effectiveness of the implemented approach with the Convolutional Long-Short Term Memory model producing optimal classification accuracy (99.4%). Moreover, this analysis results shed light on the impact of such detection mechanisms towards building resiliency into different levels of the EV charging ecosystem while allowing power grid operators to localize attacks and take further mitigation measures. Specifically, the detection mechanism of oscillatory load attacks is decentralized and created an effective alternative for operator-centric mechanisms to mitigate multi-operator and Man-in-the-Middle (MitM) oscillatory load attacks against the power grid. Additionally, the created test bed is leveraged to evaluate a distributed mitigation technique, which can be deployed on public/private charging stations to average out the impact of oscillatory load attacks while allowing the power system to recover smoothly within 1 second with minimal overhead. Finally, a security metric is createdto evaluate the resiliency of the ecosystem based on its design and architecture.

Knowledge Graph Based Large Scale Network Security Threat Detection Techniques

Abstract This paper constructs a detection technique for large-scale network security threats based on a knowledge graph, extracts the attack features of network security threats using feature template FT, and combines the CNN layer, BiLSTM layer and CRF layer to establish FT-CNN-BiLSTM-CRF large-scale network security threat detection technique. Network security threat performance evaluation experiments and multi-step attack experiments have verified the detection capability of this paper's method. The recall rate of the method built in this paper in detecting malicious data is about 62.39%, the average F1-Score for normal and malicious traffic detection is 0.7482, and the anomaly score for normal traffic detection is almost 0. The detection performance of this paper's method for multi-step network attacks is superior to that of other methods, and it is capable of detecting malicious attacks quickly. Experiments have proved that the method constructed in this paper can meet the requirements of detection capability and efficiency in large-scale network security threats and has high feasibility and application value.

Malicious Data Traceability and Throughput Optimization Using Blockchain (IOTA Smart Contracts) for Next Generation Internet of Vehicles

Malicious Data Traceability and Throughput Optimization Using Blockchain (IOTA Smart Contracts) for Next Generation Internet of Vehicles

A Probabilistic Trust Model and Control Algorithm to Protect 6G Networks against Malicious Data Injection Attacks in Edge Computing Environments

A Probabilistic Trust Model and Control Algorithm to Protect 6G Networks against Malicious Data Injection Attacks in Edge Computing Environments

The Weaponization of Artificial Intelligence in Cybersecurity: A Systematic Review

The weaponization of artificial intelligence (AI) and machine learning (ML) models in cybersecurity is a growing concern, with cybercriminal organizations and nation-states exploiting their weaknesses. The Microsoft "Tay" chatbot incident exemplifies the risks of weaponized AI, as it displayed sexist and racist behaviors due to malicious data inputs. The researcher examined 21 academic studies on AI weaponization and AI-driven cyberattacks in this systematic review. This work, a systematic review, concludes with an in-depth understanding of the scale and scope of using AI as a cyber weapon. The findings revealed that more research is necessary on weaponizing AI for offensive cybersecurity applications and the following key observations (a) a connection between AI weaponization and countermeasures, (b) AI’s role in enhancing cybersecurity defenses, (c) AI weaponization offering mitigation strategies for protecting digital assets and infrastructure, and (d) AI-driven attacks exploiting vulnerabilities, enabling automation and scalability, facilitating data poisoning and manipulation, improving social engineering, and augmenting evasion obfuscation. This study contributes to a better understanding of AI weaponization and aids researchers in synthesizing current literature on the topic.

Efficient Privacy-preserving Logistic Model With Malicious Security.

Conducting secure computations to protect against malicious adversaries is an emerging field of research. Current models designed for malicious security typically necessitate the involvement of two or more servers in an honest-majority setting. Among privacy-preserving data mining techniques, significant attention has been focused on the classification problem. Logistic regression emerges as a well-established classification model, renowned for its impressive performance. We introduce a novel matrix encryption method to build a maliciously secure logistic model. Our scheme involves only a single semi-honest server and is resilient to malicious data providers that may deviate arbitrarily from the scheme. The -transformation ensures that our scheme achieves indistinguishability (i.e., no adversary can determine, in polynomial time, which of the plaintexts corresponds to a given ciphertext in a chosen-plaintext attack). Malicious activities of data providers can be detected in the verification stage. A lossy compression method is implemented to minimize communication costs while preserving negligible degradation in accuracy. Experiments illustrate that our scheme is highly efficient to analyze large-scale datasets and achieves accuracy similar to non-private models. The proposed scheme outperforms other maliciously secure frameworks in terms of computation and communication costs.

Automated malware detection using machine learning and deep learning approaches for android applications

The popularity of mobile devices has aided in the development of the Android platform. Malware is one of the most frequent cyberattacks with its prevalence growing daily across the network. Due to these flaws, a hacker can simply access the mobile device's private data. To alleviate this issue, this paper proposes a novel Malware Attack Detection in android using deep belief NETwork (MAD-NET) which accurately detects and mitigates the malware attacks and enhances the security of the devices. Initially, the benign and malicious data's from the CICAndMal2017 datasets are given as a input to the feature extraction process. In feature extraction, the data's are classified into two types such as signature-based data and behaviour-based data. The extracted features are converted as a sequence data and fed into the classification phase. In the classification phase, the benign and malicious data's are classified by using DBN technique. After being classified, the detected malicious data's are converted into a detection report for further processing. The proposed MAD-NET approach is evaluated by using CICAndMal2017 datasets and it is simulated by using Network Simulator. The proposed MAD-NET is evaluated by using many performance metrics such as accuracy, precision, recall, and F1-score. The MAD-NET technique achieves an overall accuracy of 99.83 % for Deep Belief Network (DBN), than Artificial Neural Networks (ANN), Generative Adversarial Network (GAN) and Long Short-Term Memory Network (LSTM) technique of 93.11 %, 96.75 %, 94.42 % respectively to detect and mitigate the android devices from malware attacks.

Swarm Intelligence-Optimized Energy Management for Prolonging the Lifetime of Wireless Sensor Networks

Recent technological and industrial progress has enabled the development of small, high-performing, energy-saving, affordable sensor nodes that possess the potential to adapt, be self-aware, and self-organize. These nodes are designed for versatile communications applications. Sensor networks for sustainable development focus on the ways in which sensor network technology can enhance social development and improve living standards without causing harm to the environment or depleting natural resources. Wireless sensor networks (WSNs) offer undeniable benefits in various fields, including the military, healthcare, traffic monitoring, and remote image sensing. Given the constraints of sensor networks, varying degrees of security are necessary for these critical applications, posing difficulties in the implementation of conventional algorithms. The issue of security has emerged as a primary concern in the context of IoT and smart city applications. Sensor networks are often regarded as the fundamental building blocks of IoTs and smart cities. The WSN encompasses a routing algorithm, network strength, packet loss, energy loss, and various other intricate considerations. The WSN also addresses intricate matters such as energy usage, a proficient approach for picking cluster heads, and various other concerns. The recent growth of Wireless Sensor Networks (WSNs) has made it increasingly difficult to ensure the trustworthiness and reliability of data due to the distinct features and limitations of nodes. Hostile nodes can easily damage the integrity of the network by inserting fake and malicious data, as well as launching internal attacks. Trust-based security is employed to detect and identify rogue nodes, providing a robust and adaptable protection mechanism. Trust evaluation models are crucial security-enhancement mechanisms that enhance the reliability and collaboration of sensor nodes in wireless sensor networks. This study recommends the use of DFA UTrust, a unique trust technique, to effectively satisfy the security requirements of WSNs.

Machine learning based smart intrusion and fault identification (SIFI) in inverter based cyber-physical microgrids

The paper presents a machine learning based Smart Intrusion and Fault Identification (SIFI) method to identify the cyber-physical abnormalities in an inverter-based cyber-physical microgrid (CPM). The SIFI method utilizes an ensemble classifier (EC) to assimilate the decisions from three distinct classifiers (C4.5 decision tree, random forest (RF), and forest by penalizing attributes (FPA)). The proposed method employs a voting mechanism to classify and localize abnormal events for improved accuracy. The paper investigates the effects of cyberattacks (Denial-of-Service (DoS) and Malicious Data Injection (MDI) attacks) and physical abnormalities caused by line faults in microgrids. To train the classifiers, SIFI employs dataset with statistical attributes extracted from measurements. The renewable alternative power systemssimulation (RAPSim) software tool is utilized to model the proposed system. The effectiveness of the presented model is assessed with respect to mean value error (MVE). The efficiency of the classifier is demonstrated by comparing its performance with other classifiers in terms of the MVE. For physical abnormalities identification and localization, the approach provides enhanced outputs with the MVE of 0.157% and 0.162% correspondingly. For identifying MDI anomalies, the model provides better results with a lower error rate. For DoS anomaly detection, the model provides better classification performance with 0.136% error. The extensive empirical analysis proves that the anticipated ensemble classifier-based SIFI model yields the minimum MVE for identifying physical faults and cyberattacks. Furthermore, the effectiveness of the proposed method is identified by considering a case study in which the SIFI method outperforms well for the MDI and DoS cyber-attacks in the CPM.