Satellite communications are rapidly evolving into the Internet of Things (IoT) era, driven by the rise of CubeSats and lightweight platforms that enable faster deployment cycles, reduced launch costs, and the integration of diverse, often unregulated payloads. As these smart satellite systems become increasingly interconnected, their cybersecurity threat landscape expands, while existing defensive controls remain ill-suited to the unique constraints of satellites, including limited power, bandwidth, and computational resources.This paper introduces UNSW-IoTSAT, a first-of-its-kind dataset designed specifically for cybersecurity research in IoT-based satellite communication systems. The dataset captures realistic satellite telemetry streams within a hybrid, cyber-physical satellite environment, combined with a diverse range of satellite-oriented cyber attacks, generated using a controlled testbed that implements Consultative Committee for Space Data Systems (CCSDS) protocol standards. UNSW-IoTSAT consists of over 404,000 labelled records encompassing both normal operations and attack scenarios.Unlike existing datasets that focus on terrestrial traffic or limited satellite telemetry, UNSW-IoTSAT uniquely integrates IoT sensor data, CCSDS-compliant communication, RF-layer impairments, and realistic cyber-attack modelling. This addresses the critical gap in publicly available satellite cybersecurity datasets and provides a reproducible benchmark for developing anomaly detection algorithms, intrusion detection systems, and vulnerability assessment techniques tailored to the constraints of smart satellite networks.
We introduce AeSSCrypt, a novel authenticated encryption scheme that achieves constant cryptographic overhead through recursive Shamir-inspired polynomial transformations. Unlike conventional O(N) block cipher modes, AeSSCrypt concentrates all computational hardness into a fixed boundary of 2(t−1)+1 pseudorandom permutation (PRP) invocations, independent of message length. The scheme bootstraps a hidden recursive state via secure PRP, processes each plaintext block through degree-(t−1) polynomial evaluations over F2128, exposes only one of t shares per round, and recursively propagates the remaining evaluations as the next hidden state. We formally prove IND-CPA and IND-CCA security via hybrid-game reductions, demonstrating that adversarial advantage reduces directly to distinguishing the underlying PRP from random. A cumulative share-sum accumulator provides deterministic integrity verification, where any unauthorized ciphertext modification is detected except with negligible probability, under the assumed adversarial model and the security of the authentication mechanism. Empirical evaluation in software-only execution shows AeSSCrypt achieves 772.31 Mb/s encryption throughput—outperforming AES-GCM by 108.1% and AES-OCB by 57.7%—while maintaining negligible overhead (1.8%) when migrating from 128-bit to 256-bit security, compared to 23–40% degradation for conventional AEADs. The architecture offers quantum-upgradability through primitive substitution at cryptographic endpoints, positioning AeSSCrypt as a scalable, post-quantum-upgradable and adaptable for high-throughput and resource-constrained applications.
Audio watermarking technology plays a crucial role in digital copyright protection and content authentication. In recent years, audio watermarking methods based on deep neural networks have attracted significant attention. These methods typically consist of an encoder, a distortion simulation layer, and a decoder, enabling end-to-end training for watermark embedding and extraction. However, existing approaches still face limitations in encoder structure design, primarily reflected in the insufficient fusion between watermarks and audio features, as well as the restricted ability to model spectral details and overall structures, which affects the imperceptibility and robustness of audio watermarks. To address these issues, this paper proposes a robust audio watermarking method based on a dual-encoder U-Net and Short-Time Fourier Transform. The proposed framework constructs an embedding and extraction network for audio watermarking. Specifically, the watermark embedding network consists of a dual-encoder U-Net and a multi-scale feature fusion module, which effectively extracts and integrates features from the audio amplitude spectrogram and the watermark sequence, embedding the watermark into different spectral regions to enhance imperceptibility. Meanwhile, the watermark extraction network introduces a multi-scale fusion module that integrates local and global features through parallel convolutional paths with different receptive fields, significantly improving the watermark extraction performance. Experimental results show that the proposed method not only exhibits good imperceptibility compared to other methods on the three public datasets but also demonstrates excellent robustness against multiple attacks, with watermark extraction accuracy approaching 100% under most attacks.
With the rise of generative AI, a new generation of intelligent phishing has emerged, where attackers leverage Large Language Models (LLMs) to craft websites with high semantic plausibility, posing critical challenges to traditional defenses. Although existing LLM-powered detection approaches utilize semantic analysis, they often treat linguistic coherence as an indicator of legitimacy. Consequently, this reliance on semantic plausibility becomes a liability, creating a vulnerability that attackers adversarially exploit to bypass detection. To address this, we investigate the problem from an adversarial asymmetry perspective: while attackers can easily manipulate surface semantics at low cost, they face inherent constraints in concealing infrastructural evidence. Based on this observation, this paper proposes DEMO (Detection mechanism by integrating network Evidences and LLM-based semantic Optimization). DEMO operationalizes the adversarial asymmetry principle by explicitly incorporating network evidence auditing into semantic reasoning and employing Cyber Threat Intelligence (CTI) as knowledge anchors, thereby exposing the deep cross-layer inconsistencies in intelligent phishing. The proposed DEMO is evaluated on a real-world dataset containing 10,365 websites, achieving an F1-score of 95.97% and a recall of 99.05%, significantly outperforming state-of-the-art baseline models. Extensive experiments further demonstrate its strong robustness against adversarial perturbations and low inference latency, indicating its potential capacity for large-scale, real-time detection.
Deep Learning (DL) has proven highly effective in Side-Channel Analysis (SCA), especially against secure devices like smart cards. However, designing efficient DL models remains time-consuming and often unsuitable for new or unknown systems, as performance heavily depends on hyperparameters and architecture choices. To address this, we investigate the integration of metaheuristic algorithms into Neural Architecture Search (NAS) for SCA. We study two metaheuristic classes-population-based metaheuristics (Genetic Algorithm (GA) and Particle Swarm Optimization (PSO)) and single-solution-based metaheuristics (Simulated Annealing (SA) and Tabu Search (TS))-to automatically design and optimize DL models. Two search spaces are modeled: a smaller one for Multilayer Perceptrons (MLPs) and a larger, more complex one for Convolutional Neural Networks (CNNs), enabling evaluation under varying search complexities. Experiments on the ASCAD (Fixed Key) and CHES CTF datasets show that metaheuristic-NAS consistently outperforms traditional optimization techniques such as Random Search (RS) and Bayesian Optimization (BO). In smaller spaces, SA offers competitive results with lower execution time, while Tabu Search finds good architectures but is slower. Using the Guessing Entropy (GE) metric, in the MLP search space the SA approach achieves GE = 1 in ≈ 60 traces. In the CNN space, top configurations reach GE = 1 in ≈ 200 traces for ID leakage and ≈ 420 for the more challenging Hamming Weight leakage. Based on our experimental evaluation, metaheuristic-NAS hybrids are confirmed to be efficient and practical tools for automating DL model design in SCAs.
The Android operating system, dominating over 85% of the mobile market through open-source flexibility, suffers from intrinsic vulnerabilities. The APK(Android Package Kit) parsability and Smali code modifiability enable attackers to decompile applications via tools. This facilitates widespread repackaging-malicious actors inject payloads or tamper with functionality, redistributing counterfeit applications(apps) through third-party markets. These practices cause dual damage. Developers face code theft and revenue diversion, while users endure privacy leaks, financial fraud, and device compromise. Consequently, accurate repackaging detection has become critical. This paper reviews recent progress in repackaging detection techniques for Android applications. We first outline the fundamental characteristics of Android apps and then examine detection methods based on code analysis and resource similarity. Frequently used Android app datasets and evaluation metrics for measuring the effectiveness of repackaging detection methods are also summarized. Finally, we discuss the development trends of repackaging detection techniques and identify future research directions, with the aim of providing meaningful insights and guidance for researchers in this domain.
Advanced Metering Infrastructure (AMI) forms a crucial part of smart grids by enabling real-time monitoring and two-way communication between consumers and utility providers. However, its reliance on wireless and multi-hop communication exposes it to spoofing and identity-based attacks. In this paper, we propose an Intelligent Authentication & Intrusion Detection System (IAIDS) for securing AMI, designed to authenticate smart meters (SMs) and continuously monitor the network to detect identity based attacks in real time. It’s a novel three-phase framework that leverages the spatial correlation of RSSI patterns and machine learning to detect identity attacks. First, IAIDS estimates the expected RSSI (ERSSI) values by leveraging physical and environmental models (weather and terrain), thereby eliminating the need for predefined signal profiles. This Phase effectively reduces false alerts by accounting for normal RSSI fluctuations caused by environmental variability. Second, it utilizes unsupervised outlier detection techniques to dynamically identify potential anomalies at the local level, without relying on static thresholds. This enables each SM to adaptively detect suspicious behavior, improving detection sensitivity. Third, IAIDS confirms the anomalies through cooperative classification among neighboring SMs, enhancing precision. Evaluated on real RSSI and weather datasets, IAIDS achieves high precision and recall. This approach ensures high detection sensitivity while maintaining a low false alarm rate, making it suitable for secure, reliable, and scalable deployments within AMI network.
With the growth of virtual treatment platforms and digital mental health services, it is now more important than ever to protect sensitive user data and maintain confidentiality. However, these networks have become increasingly vulnerable to online threats such as unauthorised access, data theft, and privacy invasion. This paper introduces an architecture for digitally delivered mental health services that enhances cybersecurity through the use of artificial intelligence (AI). The use of advanced artificial intelligence algorithms to identify, stop, and respond to security threats in real-time makes virtual therapy platforms protecting patient data with our methodology a stronghold. The growing reliance on virtual health platforms for managing chronic conditions like diabetes and thyroid disorders necessitates robust cybersecurity to protect sensitive health data, especially in mental health-related therapy sessions. This report evaluates a multi-layer AI-enhanced cybersecurity architecture in a simulated environment involving 25 synthetic patient samples undergoing virtual consultations. The goal is to analyse how AI models not only secure the sessions but also ensure data integrity, monitor health outcomes, and maintain regulatory compliance. Statistical analysis of session data, anomaly detection scores, and cyber incident response metrics reveals the feasibility and efficiency of AI-driven models in safeguarding virtual consultations.
Unmanned Aerial Systems (UAS) have emerged as a transformative technology across multiple domains, including military, logistics, agriculture, and topography. However, limited attention to security during system design, combined with evolving threats and expanding attack surfaces, continues to expose UAS to significant cyber risks. Ensuring the secure operation of these systems is critical, as vulnerabilities can compromise mission success, system integrity, and public safety. To address these challenges, threat modeling is widely regarded as a fundamental practice in system security engineering. Nevertheless, the absence of methodologies explicitly developed for UAS has led practitioners to adopt approaches originally designed for IT systems. These methodologies often fail to account for the distinctive characteristics of UAS as Cyber-Physical Systems (CPS), thereby limiting their effectiveness. To bridge this gap, this study introduces the Process for Attack Simulation and Threat Analysis for Drones (PASTAD), a threat modeling methodology adapted from the PASTA framework and tailored to UAS. Built upon a layered architectural model and robust risk assessment metrics, PASTAD provides a structured, systematic, and context-aware process for identifying threats, evaluating risks, and guiding risk mitigation strategies. The methodology is demonstrated through an illustrative case study involving a fixed-wing UAS deployed in airport surveillance. This study further evaluates PASTAD, discusses its scalability and potential generalization across other UAS platforms, and outlines future directions to advance threat modeling practices and strengthen the resilience of these systems in critical applications.
Medical diagnostic systems generate sensitive patient information that requires optimal protection during transmission and storage. Image steganography provides a secure method for embedding secret data, making it imperceptible to the naked eye as part of the original image. However, applying general image steganography directly to medical images can compromise the quality of the transmitted data, and the distortions make the image hosting the secret information appear suspicious and inaccurate for medical interpretation. Steganography in medical images is in its early stages, focusing primarily on basic data-hiding techniques with limited security enhancements. This study introduces MedicalFuzzySec, a dedicated steganographic framework for concealing Electronic Patient Data (EPD) in medical images through fuzzy logic-guided difference expansion. The originality of MedicalFuzzySec lies in its adaptive embedding mechanism, which selectively identifies optimal pixel regions using fuzzy inference rules to ensure high data security with minimal impact on diagnostic image quality. MedicalFuzzySec addresses the limitations of existing approaches, including image degradation and insufficient payload handling, by offering a secure, high-fidelity solution tailored to clinical image standards. Experimental results confirm that MedicalFuzzySec consistently achieves high imperceptibility and robustness, with PSNR values ranging from 56.06 dB to 76.29 dB and SSIM values from 0.989 to 0.999, positioning it as a state-of-the-art solution for secure EPD transmission in medical systems.
In our current era, sending and receiving data securely is a significant challenge. Through data transmission, we must ensure that the data reaches the intended recipients without being compromised by hackers or unauthorized ones. Steganography is one of the best techniques for hiding data within other media without raising suspicion. The goal is to transmit data embedded covertly, making it undetectable to unauthorized. The H.264 Video compression technique provides high-quality video at low bit rates for streaming, recording, and online distribution. This research paper presents a method for hiding data within a cover media (video) using the H.264 technique, which means video steganography. The hiding is done using motion estimation in the encoder of H.264, where the bits are embedded in the motion vectors within the P and B frames for the IBBP sequence format. The hidden data is retrieved and extracted by the H.264 decoder by motion compensation in the receiver. These operations were done using MATLAB and both the secret text and the original cover media were successfully retrieved with an SNR of >34 dB.
From social and technological to biological, complex networks are found in a lot of domains. Inherently, such systems possess dynamic and intricate characteristics. One of the core problems in network study is the detection of influential nodes that decisively play a major role in the structure and functionality of such networks. This paper proposes a new method for measuring node importance by fusing the GIN of each node and ASP. Not only does it consider the inherent importance of a target node, but it also considers the overall impact of all nodes within the network. The GIN index gives a quantitative means of measuring the global influence of every node. In the paper, the ASP theory is used to assess node efficiency, considering the relative variation in the length of the shortest path over the whole network. In the present work, the authors introduce a novel centrality measure based on the harmonic mean between the GIN and ASP indices. To this end, the SIRM simulation has been conducted in order to test the diffusion of an epidemic in various network structures for evaluating the effectiveness of the proposed method. Extensive experiments on real networks have shown its efficacy and superiority compared with conventional centrality measures.
The modern technology network plays an important role. Artificial intelligence to provide a higher solution for network security. As 7G networks emerge, virtual therapy platforms have begun to take center stage, offering fast, real-time interaction and high-definition immersive experiences. While the convergence of security in health care and mobility is in many ways even more integrated into our lives, it is fraught with cybersecurity risks. This article proposes an AI-based cybersecurity framework for a 7 G-based virtual therapy platform. The framework employs advanced machine learning (ML) algorithms, predictive analytics, and adaptive threat intelligence to protect the confidentiality, integrity, and availability of data. The article covers the core building blocks like anomaly detection, secure 7 G-based virtual therapy platform authentication protocols, and privacy-preserving techniques. AI-GN gives an overview of attack mitigation using AI with a comparative analysis of existing 5G/6G frameworks and solutions and their unique aspects needed for 7G environments. This article mainly focuses on threat detection using a deep learning method, also it aims to provide a continuous authentication process. The proposed model response time is decreased by nearly 30 percent, at the same time detection rate is increased to 98.5 percent in a 7G network.
Ransomware has emerged as a formidable cybersecurity menace, targeting organizations and individuals with devastating consequences. This research paper delves into the multifaceted domain of ransomware resilience, investigating advancements in detection and mitigation strategies to combat realworld threats effectively. The study begins with a comprehensive literature review, shedding light on the evolving tactics of ransomware perpetrators and the limitations of traditional security measures. Drawing insights from diverse data sources, including incident reports and surveys, the research analyzes historical ransomware attacks, uncovering trends, attack vectors, and vulnerable sectors. In pursuit of improved detection capabilities, the paper explores cutting-edge approaches such as behavior-based analysis and machine learning algorithms. Deception methods and honeypots are investigated as additional layers of defense to divert attackers and minimize the impact of potential breaches.
In the current era, informatics systems technology is advancing at a rapid pace, and as a result, the Internet of Things (IoT) has become increasingly important to daily life in many ways. Multisensor fusion is the integration of data from several sensors/ sensing devices (i.e., smart IoT devices) to produce a more accurate and reliable representation of the environment. It is a crucial technology across numerous fields, including robotics, autonomous vehicles, smart cities, and other IoT-driven applications. The availability of several devices that serve as IoT enablers, such as smartwatches, smartphones, security cameras, and smart sensors, has led to an increase in the popularity of IoT applications compared to earlier times. In order to create a bidirectional distributed denial-of-service (DDoS) detection mechanism for an IoT-driven data fusion system, this study proposed a scheme by making use of three deep/ machine learning algorithms, K-Nearest neighbor (KNN), Gaussian Mixture Model (GMM), and Support Vector Machine (SVM). In order to identify the most efficient model against DDoS attacks that can precisely detect and discriminate DDoS from legal traffic, the KNN, GMM, SVM are tested and put into practice using SVM model for highest accuracy. An SDN-specific data set created with Mini Net emulator involves designing a network topology, generating traffic, and collecting data to evaluate SDN applications and controllers. Confusion Matrix is used to test and evaluate the three models in relation to four widely-used criteria: accuracy, precision, recall, and F-measure. Network simulation is used to analyze malicious traffic, which consists of a combination of ICMP, UDP Flood, and TCP Syn attack, as well as benign TCP, UDP, and ICMP traffic.
An emerging application of such new technologies is in urban development, with cities increasingly utilizing them to address social, environmental, and urban issues. IoT has paved the way for Smart Cities, while AI-fueled big data has revolutionized progressive urbanization. However, initiatives to promote technology must be balanced by principles of sustainability and livability. As deep learning has advanced rapidly, creating increasingly sophisticated technologies has led to highly complex — and often opaque — models that can be difficult to interpret. It becomes increasingly difficult to establish trust and maintain transparency when decision-making systems are based on such opaque and complex structures. This article explores the urban promise of AI and presents a new framework infusion of AI into cityscapes. The new direction is socially oriented through the inclusion of elements such as values, urban metabolism, and governance. A systematic review of machine-learning applications in cybersecurity also discusses the importance of explainability for overcoming the challenges it entails. The importance of assuring the explainability, interpretability, and intelligibility of autonomous systems will also be part of this discussion, especially in the context of developing smart cities using AI-based technologies.
Phishing attacks represent a growing threat to online users and software developers, necessitating the development of advanced detection strategies. This study proposes a hybrid framework that integrates convolutional neural networks (CNN) for feature extraction and support vector machines (SVM) for classification, with the SVM optimized using the grey wolf optimizer (GWO). The CNN component is responsible for capturing complex and discriminative patterns from website data, enabling more effective differentiation between phishing and legitimate websites. Hyperparameter tuning via GWO enhances the classification performance of the SVM by generating an optimal decision boundary. Evaluation was conducted using established datasets, including those from Kaggle, the UCI Machine Learning Repository, Phishtank, 5000 Best Websites, and Alexa. Experimental results show that the CNN–SVM model, with GWO optimization, achieved an accuracy of 99.18 %, indicating its practical utility in phishing detection applications. The findings suggest that the proposed framework, supported by additional security mechanisms, contributes to a reduction in false positives while maintaining reliable detection of phishing threats.
The fusion of image steganography with blockchain technology have brought about novel prospects for safe and decentralized authentication systems in recent years. In this paper, we propose a blockchain-based authentication system that makes use of eXtended Difference of Gaussians (xDoG) edge detection in steganography domain. Here, the xDoG edge detection captures both major and subtle edges for a more detailed and nuanced edge map. Other parameters viz. edge thickness, contrast and brightness etc. may be fine-tuned, making it more prominent. The principle behind embedding the secret image inside the cover image is to conceal more secret bits in the edge pixels and fewer bits in the non-edge pixels. The Stego-image obtained through the process is conveyed to the intended recipient using a public channel. The sender computes a hash value from the Stego-image and records the same in the blockchain network which makes illegal access and data tampering more stimulating. Blockchain's intrinsic immutability and decentralized structure are exploited to verify the authenticity and integrity of the Stego-image, allows the receiver to extract out the hidden content and, thereby firming resistance against attacks. Simulation result ensures that the proposed method attain an average payload of up to 3.43 bpp. The average PSNR in this trial consistently stays above 34 dB, and the SSIM never falls below 0.93, guaranteeing that the Stego-images' image quality is both very acceptable and better than that of the state-of-the-art techniques. To confirm the findings, the StegExpose and SR-Net are also used to assess the security of the Stego-image.
Quantum communication is a rapidly evolving field that leverages quantum physics to enable secure and efficient information exchange. A cornerstone of quantum cryptography is Quantum Key Distribution (QKD), a secure key exchange mechanism that ensures the confidentiality of transmitted data by employing the no-cloning theorem and the uncertainty principle of quantum mechanics. This study explores key aspects of quantum networks and communication, including quantum states, QKD, and quantum cryptographic protocols, with a focus on photon polarization states and entangled qubits as fundamental building blocks of quantum information. Additionally, two well-established quantum cryptographic protocols, BB84 and E91, are analyzed for their principles and advantages in secure communication. However, challenges such as quantum state loss in communication channels hinder the long-distance transmission of quantum information. To address these issues, error detection, measurement, and correction techniques are investigated, with quantum error correction methods playing a crucial role in mitigating noise and imperfections, ensuring accurate quantum information transmission, and enhancing the overall efficiency of quantum communication systems. Beyond conventional networks, quantum communication holds vast potential for cybersecurity applications, paving the way for next-generation secure communication frameworks.
