Abstract : The article explores emerging security threats in sixth-generation (6G) networks and evaluates the applicability of machine learning (ML) models for threat detection and prevention. It provides a typology of common attack vectors, matches them with relevant ML techniques, and emphasizes the importance of selecting context-specific algorithms. The paper further discusses architectural considerations for integrating ML into the distributed infrastructure of 6G and presents a comprehensive evaluation framework based on technical and operational metrics. The study concludes that adaptive, interpretable, and energy-efficient security systems are essential for maintaining resilience and enabling localized analysis in next-generation networks.

Abstract : This article discusses the concept of adaptive quantum-resistant encryption (AQRE), designed to protect data in intelligent information systems vulnerable to quantum computing threats. The main principles of implementing AQRE are described, along with its role in various application fields such as autonomous transportation systems, medical platforms, and financial networks. The interaction of key system components, such as the context evaluation, threat prediction, encryption selection modules, cryptographic container, and feedback module, is also explored. The article analyzes the effectiveness of adaptive encryption in real-world scenarios and discusses the future development of technologies, including integration with machine learning methods and hybrid computational architectures. The importance of an adaptive approach for ensuring data security in the face of constantly evolving threats is emphasized.

Abstract : This article explores secure multi-party computation (SMPC) as a foundational cryptographic approach for performing collaborative analytics on sensitive big data without compromising privacy. The study analyzes the architectural components, protocol mechanisms, and practical considerations for integrating SMPC into large-scale analytical systems. Key focus areas include data representation, secure aggregation, performance optimization, and interoperability with machine learning workflows. Through illustrative examples and technical evaluation, the paper highlights current limitations and emerging solutions for scalable, privacy-preserving computation. The findings offer insights into designing secure analytics pipelines suitable for real-world deployment across regulated and distributed environments.

Abstract : Interaction models in intelligent sensor networks (ISNs) form the basis for autonomous communication and coordination in industrial internet of things (IIoT) systems. The analysis focuses on topological structures, hierarchical communication layers, synchronization strategies, and decentralized behavior control. Core challenges related to interoperability, temporal consistency, and field-level integration are discussed alongside technical patterns for achieving scalable and resilient performance. The findings contribute to the development of robust ISN infrastructures capable of operating under the complexity of modern industrial environments.

Abstract : This paper presents a comparative evaluation of data sharding strategies used in distributed ledger systems. The analysis explores partitioning methods, cross-shard transaction protocols, storage architectures, and security implications associated with fragmenting ledger state. Particular attention is given to performance trade-offs, consistency management, and resistance to targeted attacks in partitioned networks. The findings offer practical insight into how different sharding approaches influence system scalability, responsiveness, and reliability. The study serves as a foundation for future design choices in high-performance ledger infrastructures.

Abstract : Hybrid forecasting systems have become essential for anticipating dynamic resource demands in cloud computing. By integrating machine learning, time-series modeling, and adaptive mechanisms, these systems enable accurate load predictions across heterogeneous workloads and fluctuating usage patterns. The study explores the design and evaluation of such models, highlighting architectural considerations, empirical trade-offs, and real-time deployment strategies. Results from comparative experiments demonstrate the effectiveness of hybrid approaches in reducing forecasting error and improving provisioning efficiency. Emphasis is placed on system responsiveness, model adaptability, and performance under operational constraints.

Abstract : Blockchain-based digital identity systems are reshaping how individuals and institutions manage identity credentials across jurisdictions. By leveraging decentralized identifiers, verifiable credentials, and distributed trust models, these systems enhance privacy, data control, and interoperability. This paper investigates the architectural foundations, governance mechanisms, and institutional challenges associated with cross-border deployment. Key emphasis is placed on privacy-preserving strategies, regulatory alignment, and multistakeholder trust frameworks. The study highlights that while technical standards provide a solid base, scalable adoption depends on legal harmonization and institutional integration.

Abstract : The integration of autonomous intelligent agents into decision support systems enhances the capacity of critical infrastructure to operate reliably in dynamic and uncertain environments. These agents provide essential functions such as real-time monitoring, adaptive response, distributed coordination, and learning-based optimization. The article examines the functional architecture, communication patterns, and implementation strategies of autonomous agents across different infrastructure domains. Through architectural modeling, decision logic representation, and analysis of scalability and fault tolerance, the study demonstrates how agents support resilient, decentralized decision-making. Particular attention is given to layered integration, enabling agents to function effectively at sensing, control, coordination, and strategic levels. The findings contribute to the development of intelligent, explainable, and adaptable decision support frameworks for infrastructure resilience.

Abstract : Resource allocation in edge computing systems must account for the strict timing, reliability, and locality requirements of real-time applications. This study provides a structured analysis of edge architectures, application classes, and optimization criteria for task distribution. Several algorithmic strategies are outlined, including priority-based, delay-aware, and predictive methods, with particular attention to network dynamics and heterogeneity. Emphasis is placed on the role of adaptive and hybrid models that enhance resilience and performance in dynamic computing environments.

Abstract : Modern power grids generate large-scale, heterogeneous data that require advanced analytical approaches for effective failure prediction and risk mitigation. This article explores the integration of Big Data technologies and deep learning models to enable predictive analytics across critical power infrastructure. A detailed analysis of model architectures-including LSTM, GRU, CNN, and Transformer-is provided, with comparisons based on temporal modeling capabilities, computational efficiency, noise tolerance, and real-time applicability. The study further examines implementation scenarios, system integration challenges, and reliability considerations. Prototypes and evaluation metrics are discussed to support practical adoption. The findings highlight the importance of designing adaptive, explainable, and scalable solutions that align with the complexity and safety demands of real-world grid environments.