Annals of Telecommunications最新文献

The Internet of drones (IoD) paradigm is an emerging technology that has gained significant attention from the research community in the recent years. Drone networks are now widely utilized in various sectors and industries. However, like all emerging technologies, these networks face numerous challenges, with security being the most critical. The wireless nature of drone communication makes them vulnerable to a variety of cyber attacks. In addition, the limited resources of drones pose challenges in implementing effective security solutions. This paper proposes a new approach to classifying drone cyber-attacks based on their similarities. We introduce a new step-by-step framework model to classify cyber attacks. Then, we highlight the existence of commonalities between different cyber attacks that can occur in drone network systems. This classification will serve as the foundation for the development of future unified solutions capable of efficiently mitigating multiple attack types simultaneously.

Large Language Models (LLMs) are a significant leap in Artificial Intelligence (AI), providing tremendous capabilities for understanding and creating human-like language. LLMs offer significant advantages in automating and enhancing cybersecurity practices, but their deployment in sensitive domains like cybersecurity, ethical hacking, and digital forensics is a challenge. LLMs improve cybersecurity, digital forensics, and ethical hacking by automating processes such as reconnaissance and advanced attack simulation to confront new digital threats. This study examines LLMs’ strengths and weaknesses and proposes mitigation strategies, such as encryption, adversarial training, and strict data governance, to ensure their safe and effective integration in high-stakes situations. The results highlight the necessity of ongoing improvement, supervision, and ethical concerns to fully utilise LLM while reducing dangers and guaranteeing its appropriate deployment.

Software-defined networking (SDN) is a pioneering paradigm that adds a new level of management and programmability to computer networks. Despite its benefits, SDN opens the door for new attacks against the network by targeting different SDN components, especially when they need to interact with each other through application programming interfaces (APIs). In this work, we focus on protecting those APIs by proposing SDN-API-Sec-Fast, a cross-domain access control method based on the key characteristics of blockchain technology. We enhance our previous work SDN-API-Sec by introducing SDN-API-Sec-Fast, which combines conflict detection with read-only mode to achieve better performance. We investigate the performance of the blockchain side and the overall authorization process.

Smart interconnected devices belonging to the Internet of Things ecosystem are resource-constrained in terms of hardware and software. They are also prime attack targets for malicious parties. Although there has been an extensive exploration of attack detection methods rooted in machine learning, such approaches necessitate high processing overhead, which is ill-suited for devices of modest processing capabilities. Furthermore, machine learning algorithms are opaque black boxes. Therefore, we present a novel hybrid approach to detect distributed denial-of-service attacks using fuzzy cognitive maps paired with machine learning feature selection. Our approach incorporates contextual information (features) drawn from network packets. We utilize feature selection methods to compute the weights of the features. The weights capture the influence of each input feature on the target output feature that determines the classification of any packet as malicious or benign. The features and weights are used to construct a fuzzy cognitive map for each type of attack. The fuzzy cognitive map is then used to train and test the dataset. We also auto-compute a threshold value that allows our model to classify a packet as malicious or benign. Our model performs best using the weights computed by two particular statistical feature selection algorithms, namely, SelectKBest-Classification and SelectKBest Chi-squared, combined with FCM. Our experiments show that this hybrid approach is simple, reliable, and transparent with a low memory footprint, and therefore well-suited for devices with limited resources.

Cyber-physical energy systems (CPES) are the core components of an energy system. A CPES is characterized by a high level of complexity and uncertainty, while ensuring a more resilient and efficient distributed energy system, including robust energy security and economic viability. This work presents a novel and effective approach for anomaly detection in CPES systems using advanced tree-based machine learning models. The proposed approach is centralized with a minimal set of features that achieve a maximum detection rate and system performance. The results emphasize the effectiveness of ML methods in this domain, showcasing their ability to handle complex, high-dimensional datasets while providing interpretability and efficiency. The study also highlights associated challenges such as scalability and adversarial resilience, proposing future research directions such as hybrid models and interpretable AI for real-world deployment.

Secure and efficient electronic ticket (e-ticket) management in the entertainment industry is important for both the compliance of service-level agreements and the pride of those involved. Despite its numerous benefits, the ease of duplication, fraud, and illegal resale poses several issues in the industry, which can adversely impact event organizers and legitimate buyers. Furthermore, security breaches could compromise buyers’ personal and financial information. An e-ticketing system based on blockchain is implemented, mainly designed for managing raffles. This system issues e-tickets recorded in encrypted form in a blockchain with multiple blocks. The system digitally signs each ticket, links it to the buyer, records the data on-chain, and conducts draws using secure cryptographic randomness. At the end of the process, the system provides an option to download an audit chain that facilitates the validation and audit of the draws. We used the agile Scrum methodology and Java to develop the e-ticket system, implementing the blockchain with a PostgreSQL database and a custom hash function. We store each block in an unstructured JSON-type field. To optimize response times in processing the proof-of-work (PoW) blockchain, a Java socket server was implemented to manage mining pools and distribute the load among miners. We organized the load into groups by applying the K-means algorithm to historical performance data. This classification enabled the formation of efficient groups that accelerated PoW resolution, enhancing the security and performance of raffle creation, signing, and execution, particularly under increased hash difficulty. Each miner received rewards based on their contribution, fostering community participation and encouraging continuous improvement in individual performance. We evaluated the system with an average block time of 30 s and a cost of ($ )0.005, demonstrating economic viability for popular raffles with affordable ticket prices. In security tests, 10% of counterfeit tickets were successfully injected and rejected, demonstrating tamper resistance. These results confirm the feasibility of open, secure, and auditable raffles through blockchain verification, strengthening trust in public systems based on distributed technologies.

In this study, we introduce a quantum-resistant and lightweight authenticated key agreement protocol tailored for the Multimedia Internet of Things (MIoT) environment. MIoT devices gather and send sensitive multimedia data to cloud servers, which means that secure mutual authentication and quick session key creation are needed. To accomplish feasible lattice-based key exchange, our scheme uses a reconciliation technique and the Ring Learning With Errors (RLWE) assumption. While informal analysis demonstrates resilience to common threats like replay and impersonation, the random oracle model formally proves security. The robustness of the protocol against a variety of active attacks is confirmed through additional verification using the AVISPA tool. Further performance analysis shows that the suggested method clearly improves post-quantum security and computational efficiency when compared to current authenticated key agreement schemes, while facing slightly higher communication costs to ensure stronger security. In general, this work strengthens the provable security of smart devices enabled by the MIoT. Future research will investigate more extensive MIoT applications and energy-efficient optimizations.

Distributed policy negotiation is a key issue in modern manufacturing landscapes. However, existing solutions struggle with ensuring efficient policy negotiation and enforcement across entities while maintaining device authentication and trust in dynamic, cross-organizational environments. Specifically, the reliance on distributed architectures introduces challenges related to scalability, data fairness, and implementation complexity, limiting its applicability in large-scale industrial networks. This paper provides an overview of foundational concepts critical to understanding the proposed distributed policy consensus with blockchain integration: the proposed architecture leverages the decentralized and immutable nature of blockchain to achieve secure and transparent consensus among the distributed policy engines (PE), addressing challenges in policy negotiation within a distributed zero trust architecture (ZTA). The proposed methodology is applied in the context of a multinational supply chain network, involving various stakeholders such as manufacturers, suppliers, logistics providers, distributors, and retailers. The methodology provides a scalable foundation for advancing secure cross-company interactions in increasingly interconnected systems.

The application of digital models in industries is recently being used as a means to enhance the operational efficiency of cyber-physical systems (CPSs). However, this comes with challenges bordering mostly around the detection of cyber-attacks using replication as a means. This study considers a synthetic digital object dataset and its equivalent physical object for intrusion detection using principal component analysis (PCA) and an autoencoder for dimensionality reduction. Application of random forest and XGBoost machine learning on both objects is applied to compare the performance on the two objects. To fill a methodological gap, the study implemented the developed model on the digital object dataset and its physical object equivalent to assess intrusion detection capability. The study established that the application of the developed random forest model on the use case digital object dataset varies slightly from its physical equivalent as assessed using standard performance metrics, though revealing very good performance for both objects, which validates the digital object and its applicability in the real world.