Horizon Scan of Emerging Issues at the Intersection of National Security, Artificial Intelligence, and Human Performance Enhancement.

Next-generation AI for advanced threat detection and security enhancement in DNS over HTTPS

RIS/IRS-assisted physical layer security enhancement scheme and its application to SCMA systems

Free-space optical (FSO) communication is a promising key technology for large bandwidth, high data rate and cost effective data transmission. However, FSO systems experiences crucial challenges under atmospheric turbulence, pointing errors and eavesdropping threats. The proposed machine learning framework uses generative adversarial networks (GANs) for eavesdropping threats and malicious intrusions to improve the security. The GAN based framework influences a generative model to simulate attacks, such as eavesdropping and jamming, whereas the adversarial model learns to identify and mitigate these threats in real time. By continuously adapting these strategies, the GAN framework enhances the robustness of the FSO communication link. Experimental results show that the proposed framework minimizes interception threats.

Cyber-Physical Systems (CPS) implement critical infrastructures, in which physical objects interact with services of the cyber domain consequently building a heterogeneous System-of-Systems (SoS). Although this marriage extends the functionality of traditionally closed systems, it also introduces a variety of challenges – especially for engineers. One critical aspect relates to establishing and sustaining a sufficient level of security, as exploited vulnerabilities may cause severe effects, either towards involved humans or sensitive information. Consequently, engineers must be aware of potential security-related threats and vulnerabilities. To this end, threat models are usually used to identify such weaknesses within a specification. However, existing solutions may not be able to comprehensively identify all threats in a CPS-like environment, as they often do not consider all relevant dependencies between interacting systems on a SoS level. To address this gap, we have elaborated a methodology – called Semantic Threat Model (STM), which can identify and evaluate potential threats towards a given CPS specification. In detail, the framework focuses on the semantic relationships and side effects between security objects, e.g., attacks, and the actual specification of the CPS. In contrast to existing solutions, STM takes a SoS point of view, while analyzing semantic data to gain a comprehensive view on security. The quantitative output of the method can then be used to identify the most severe attacks or to point out necessary security enhancements. We highlight the usage and benefits of the STM in the form of a case study in the domain of intelligent transportation systems.

Traditional network security mechanisms, including firewalls, intrusion detection systems, and cryptographic schemes, have become insufficient in combating increasingly sophisticated cyberattacks. The emergence of decentralized and large-scale network environments, such as the Internet of Things (IoT) and cloud computing, has further complicated the protection of data integrity, privacy, and trust management. In this context, blockchain technology provides a new paradigm for enhancing network security through decentralization, immutability, and cryptographic consensus mechanisms. This paper explores the achievements and future prospects of blockchain applications in network security based on an extensive review of recent studies, which are mainly published in IEEE, Springer, and Elsevier, as well as experimental case studies from real-world blockchain-based security frameworks. It introduces the principles of blockchain and the challenges inherent in current network security systems. Then, it investigates how blockchain has been effectively utilized in distributed denial-of-service (DDoS) mitigation, threat intelligence sharing, and secure data sharing in IoT environments. Existing challenges of blockchain-based applications are also discussed, including scalability, interoperability, and energy efficiency. Furthermore, it evaluates potential directions for future research and development. The findings suggest that integrating blockchain into network security infrastructure can significantly improve transparency, resilience, and trustworthiness in digital ecosystems.

The deep learning technique has emerged as an exemplary model for managing the Artificial Intelligence-based Blockchain framework with technological enhancements to guarantee reliable data through the consensus procedure. The deep learning-enabled blockchain transaction model has involved the development of security to solve the problems of confidentiality and data anonymity. The Hybrid techniques of the Blockchain with the Deep Learning technique are proposed to generate enhanced data durability and its propagation through the enhanced convolutional temporal network (EnCTN) for transaction analysis in a blockchain-enabled Auto Encoder technique. The sliding window extraction technique is used to extract information from a particular window size to evaluate the needed input values from the temporal series. The dilated Convolution is used to capture the long-range dependencies. The proposed technique is implemented in the Ethereum environment using Python, and experimental results show that it has produced an improved performance than the relevant technique in several performance parameters. The anomaly classification accuracy is improved than the relevant technique and it is evaluated using the NSL-KDD dataset. The proposed framework delivers an efficient solution for the real-world anomaly detection application while accurate discovery of temporal anomalies and computational efficiency is enhanced.

Proctoring Online Assessments: Enhancing Security and Academic Integrity in Open Distance eLearning

In response to the growing demand for assistive devices under the backdrop of population aging, this paper proposes an innovative collaborative navigation algorithm integrating smart wheelchairs and AI glasses based on multimodal data fusion. The algorithm optimizes a closed-loop interaction among the environmentuserdevice triad. By integrating the wheelchairs autonomous navigation capabilities with the AI glasses advanced environmental perception and real-time interaction functions, it significantly enhances system safety, autonomy, and user experience. The proposed approach employs a hybrid model combining Deep Belief Networks (DBN) and Stacked Autoencoders (SAE) to model and fuse multimodal data. Further integration with Convolutional Neural Networks (CNN) and Recurrent Neural Networks (RNN) enables improved system performance. Experimental comparisons in complex dynamic environments demonstrate that this method outperforms traditional Kalman filterbased fusion techniques. User survey results indicate a high level of acceptance and willingness to adopt this system among the target population. To promote its broader application, future research will focus on algorithm optimization, outdoor performance testing, and the enhancement of data security and privacy protection mechanisms.

Wireless Sensor Networks (WSNs) play a critical role in Internet of Things (IoT) applications such as environmental monitoring, healthcare, and industrial automation. However, WSNs face severe challenges, including limited energy resources, unreliable communication links, and growing exposure to security threats. Software-Defined Networking (SDN) provides centralized programmability and global visibility, enabling dynamic traffic management and security enforcement. In this work, we propose an SDN-based WSN solution that integrates traffic analysis for anomaly detection and security policy enforcement for attack mitigation. The SDN controller collects flow statistics from sensor nodes, analyzes traffic patterns, and enforces security policies through dynamic flow rule installation. We implemented the proposed architecture in OMNeT++, and preliminary results show that the framework can detect abnormal flows and mitigate them by blocking malicious nodes.

As cyber threats continue to increase in sophistication, the security of Operational Technology (OT) environments has become a paramount priority for organizations in various sectors. OT systems, including Industrial Control Systems (ICS) and Supervisory Control and Data Acquisition (SCADA) systems, are vital in the functioning of critical infrastructure but often lack robust security due to legacy security weaknesses. This research discusses the implementation of Multi-Factor Authentication (MFA) as a baseline strategy for enhancing the security of such systems. We recognize the distinct cybersecurity issues of OT environments, especially the use of legacy hardware that does not have contemporary security mechanisms. By suggesting an extensive framework for implementing MFA, this research offers a multi-layered system that incorporates knowledge-based, possession-based, and biometric authentication techniques. We further stress the need for role-based access control, ongoing monitoring, and user training to enhance security mechanisms. Using case studies and real-world examples, we show how MFA can be used to counter unauthorized access and increase system resilience. We present actionable recommendations for organizations wishing to deploy MFA, including mitigation strategies that reduce identified vulnerabilities to acceptable levels for critical infrastructure as a foundation of their cybersecurity approach, with the ultimate goal of safeguarding critical infrastructure and sensitive information in a hyper-connected world. Our research not only adds to the body of knowledge but also acts as a guide to deploying stringent security controls in OT networks.

This article provides a comprehensive analysis of the application of artificial intelligence (AI) technologies in the context of the information warfare conducted by the russia against Ukraine amid the full-scale armed aggression that began in 2022. The study highlights that the informational dimension of modern military conflict has become a critical instrument of full scale invasion with next-generation digital technologies that are based on artificial intelligence. The research focuses on the mechanisms of information influence implemented through advanced machine learning algorithms, generative language models, neural networks, deepfake technologies, and large-scale data analysis tools. Owing to its high adaptability and capacity to simulate human speech and behavior, AI enables to create a highly convincing fake content, to manipulate public opinion, emotional influence, and to shape a distorted perception of reality both domestically and internationally. The primary objectives of these actions include not only the population disorientation but also the erosion of trust in Ukrainian state institutions and international partners, as well as the discrediting of Ukraine’s resistance efforts. The article further examines the active use of these technologies for generating fake news, manipulative content, fake videos, and images that play a significant role in disinformation campaigns. Such campaigns are aimed to demoralize Ukrainian society, induce panic, and weaken Ukraine’s international support. Special attention is devoted to the analysis of bot networks and automated digital dissemination systems that facilitate the mass propagation of destructive content. The study also explores how AI is employed for personalized information influence targeting specific audience groups. The given facts of AI-driven digital attacks on Ukraine’s information space within social networks, messaging platforms, and online media are provided. In conclusion, the article outlines strategies for countering these threats, emphasizing the importance of cross-sectoral cooperation, enhancement of digital security, implementation of robust information verification mechanisms, and improvement of digital and media literacy among the population. It is concluded that, despite its destructive potential in the hands of aggressors, AI can serve as an effective strategic tool for information security in democratic states confronting modern information warfare.

This analysis primarily examines the compatibility of IoT security and DL in offering attack prevention services for IoT devices. Consequently. The field of Internet of Things (IoT) is seeing significant growth. Consequently, there is an increasing amount of data being exchanged between cloud technologies and wireless networks to ensure smooth data flow among linked devices. Simultaneously, the Internet of Things (IoT) proves to be an exceedingly susceptible ecosystem that is susceptible to a wide range of threats. These entities has the capacity to cause extensive destruction to a country and also pose the most significant immediate economic threat. This survey essay aims to thoroughly examine the integration of 5G, IoT, and security. It is crucial to prioritise the requirement for security systems. The introduction discusses the infiltration of web vulnerabilities into devices, based on the provided information about attacks and threats. Subsequently, it demonstrates the crucial requirement for efficient security protocols. Consequently, it proceeds by examining the interdependence between DL and security, highlighting the security enhancements provided by DL to IoT. This study examines several endorsements of the DL-based technique for detecting many assaults across a diverse range of IoT environments, such as DSOS, DDoS, probing, user-to-root, remote-to-local, botnet, spoofing, and man-in-the-middle attacks, among others. In addition, the work provides a comprehensive description of deep learning techniques and the difficulties that arise when integrating deep learning-based security solutions in the context of the Internet of Things (IoT). Furthermore, the effectiveness of (DL) in enhancing IoT security has been confirmed by detailed case studies and real-life experiences. The poll also conducts a thorough analysis of security requirements in 5G IoT networks, emphasising the importance of understanding the vulnerabilities that may arise during times of crisis and chaos. This survey focuses on the comprehensive analysis of the delicate relationship between IoT, DL, and security. It provides numerous options for gaining a deeper knowledge and effectively managing the difficult security challenges in the IoT ecosystem.

Security enhancement of three POMs based interference algorithm using elliptic curve cryptography

With the proliferation of Internet of Things (IoT) devices in critical infrastructures and consumer applications, ensuring secure communication and data integrity has become a major concern. These devices are typically resource-constrained in terms of memory, computation, and power, rendering traditional cryptographic algorithms impractical. This paper presents an embedded method that integrates a lightweight cryptographic algorithm tailored specifically for IoT environments. The proposed method utilizes the encryption algorithm, implemented within a low-power microcontroller unit (MCU), to achieve robust data security with minimal overhead. The implementation was tested on an ARM Cortex-M0 platform, and performance metrics such as encryption time, memory footprint, and energy consumption were evaluated. Compared to conventional methods like AES, this method demonstrates a 40% reduction in power usage and a 35% improvement in execution speed. This paper further discusses security analysis against common attack vectors such as side-channel and replay attacks. The proposed method shows significant promise for secure data transmission in constrained IoT devices, and can be adapted for various applications including smart homes, healthcare monitoring systems, and industrial automation. Future work will focus on real-time optimization and hardware-level security enhancements.

Machine learning (ML) has become a key ingredient in revolutionizing the physical layer security of next-generation devices across Industry 4.0, healthcare, and communication networks. Many conventional and unconventional communication architectures now incorporate ML algorithms for performance and security enhancement. In this study, we propose an unconventional, high-data-rate, machine-learning-driven, secure random communication system (HDR-MLRCS). Instead of utilizing traditional static methods to encrypt and decrypt alpha-stable (α-stable) noise as a random carrier, we integrated several ML algorithms to convey binary information to the intended receivers covertly. A support vector machine-aided receiver (SVM-R), Naïve Bayes-aided receiver (NB-R), k-Nearest Neighbor-aided receiver (kNN-R), and decision tree-aided receiver (DT-R) were integrated into a single architecture to provide an accelerated data rate with robust security. All intended receivers were pre-trained on a restricted-access dataset (R-𝓓) and exploited a static key—the pulse length—to generate and successfully classify α-stable noise samples to extract hidden binary digits. We demonstrated the performance of the proposed HDR-MLRCS by simulating 4-bit and 1000-bit transmissions (including bit error rates and confusion matrices) from the perspectives of the intended receivers and the eavesdropper receiver (E-R). The significance of the HDR-MLRCS lies in its significantly higher data rates compared to previously proposed counterparts using static receivers. At the same time, the SVM-R consistently outperformed all other considered intended receivers. Moreover, the decisive failure of E-R ensures the architecture’s resistance to possible interception of communications. The fusion of high data throughput and robustness, enabled by the utilization of ML and α-stable noise as a random carrier, highlights the suitability of HDR-MLRCS for future secure communication infrastructures.

Secure and efficient data sharing in Industrial Internet of Things (IIoT) is a continuous difficulty due to the limits of static proxy node selection, centralized designs, and the lack of agility in dynamic situations. Traditional systems often suffer from excessive latency, single points of failure, tight access control, and vulnerability to targeted attacks. To address these limitations, we offer BDEQ (Blockchain-based Dynamic Edge Q-learning), a novel framework combining blockchain smart contracts and deep Q-learning for real-time, trust-aware proxy node selection. Unlike static systems, BDEQ's reinforcement learning agent dynamically selects appropriate edge nodes based on performance, resource availability, and trust criteria. This ensures secure access control, decentralized auditing, and resilience to security attacks. In a simulated gas-industry IIoT context, BDEQ lowered data access latency by 35% and boosted throughput by 28% over baseline approaches while giving greater resilience to attacks. These results validate BDEQ's relevance to next-generation IIoT contexts needing adaptive, decentralized, and secure data sharing.

Blockchain technology relies on cryptographic mechanisms for transaction security and data integrity. However, the growing computational complexity, high transaction costs, and scalability issues pose significant challenges to blockchain adoption. Traditional cryptographic methods—such as hashing, key generation, encryption, and decryption—introduce excessive computational overhead, leading to energy inefficiencies and increased latency. This research proposes an optimization-driven crypto analysis framework that integrates metaheuristic algorithms, combinatorial optimization, reinforcement learning, and game theory to enhance the efficiency and security of blockchain cryptographic processes. The framework focuses on optimized cryptographic computation, gas fee reduction in smart contracts, security enhancement against cryptanalysis, and improved scalability of consensus mechanisms. Experimental evaluations demonstrate up to 39.4\% reduction in cryptographic execution time, 29.4\% savings in smart contract gas fees, and 33.3\% improvement in decentralization of Proof-of-Stake validators. These results validate the effectiveness of the proposed framework in achieving secure, scalable, and cost-efficient blockchain operations.

Mobile Financial Services (MFS) has changed the payment landscape in Bangladesh, bringing financial inclusion along with convenience. This study specifically explores adoption, usage patterns, scope, benefits, challenges and expectations for the future of mobile payments, focusing on bKash, Rocket, Nagad and uPay. A quantitative questionnaire-based survey was conducted with 122 participants from Sylhet City. Results reflect strong adoption (91.8%), led overwhelmingly by bKash (90.2%). Most users use mobile payments to send money, buy online and pay bills, with high transaction frequency and monthly volume. It offers key benefits such as 24/7 accessibility, time savings, convenience, and SME support, but also challenges such as scam vulnerability, fraud risk, and agent liquidity shortages. Non-users overwhelmingly use cash due to trust and complexity concerns, though low fees as well as improved security are mentioned most frequently as appealing the most. The results point to the importance of regulatory assistance, security enhancements, and financial education for developing sustainable and inclusive growth. The findings of this study provide valuable implications for policymakers, financial institutions and fintech companies to promote mobile payment adoption in Bangladesh and support the country’s move towards a cash-light society.

ABSTRACT With the rapid development of the global aquaculture industry, smart aquaculture has become increasingly essential for maintaining a stable supply of aquatic products. Traditional machine learning models frequently exhibit limited semantic understanding and insufficient scalability in complex, high‐dimensional environments. In contrast, multimodal large models integrate multi‐source information, significantly enhancing semantic depth, environmental adaptability, and natural interaction. This review explores recent research progress and future application prospects in the transition from traditional machine learning models to multimodal large models in aquaculture. The literature search was conducted using the Web of Science, Scopus, and Google Scholar databases. Based on literature from the past decade, this review summarises the evolution from traditional machine learning models to multimodal large models, emphasizing their recent advantages in integrating multi‐source data such as images, sensor readings, videos, audio, and text. By analyzing the current status and limitations of traditional machine learning in water quality monitoring, biomass estimation, behavior analysis, and health assessment, this review offers a comparative perspective on the potential of multimodal large models in critical aquaculture domains including sustainable water management, precision feeding, abnormal behavior detection, disease diagnosis and prevention, intelligent breeding, energy management, and embodied intelligent robotics. Multimodal large models still face several challenges, including data acquisition in aquaculture, enhancement of model performance, security and data governance in digital twin systems, and the advancement of industrial technologies. The review suggests that multimodal large models are poised to become key technological enablers in advancing aquaculture toward greater precision, sustainability, and automation.