International Journal of Communication Networks and Information Security (IJCNIS)

Emerging AIoT Technologies for Efficient Data Collection and Decision Making in Smart Farming

Omkar Singh, Vinoth R, Abhilasha Singh, Navanendra Singh — Thu, 01 Jan 2026 00:00:00 +0000

Traditional farming is being transformed into intelligent, data-driven agriculture by the confluence of Artificial Intelligence (AI) and the Internet of Things (IoT) into the AIoT paradigm. In order to increase agricultural output, optimize resource use, lessen environmental effects, and improve farmer decision-making, smart farming uses real-time data from distributed sensors, automated systems, and predictive models. The growing use of AIoT technology in agriculture addresses global issues such as resource limitations, population expansion, labor problems, and climate change. Wireless sensor networks (WSNs), drones and autonomous vehicles, edge computing, cloud analytics, and machine learning algorithms for predictive insights are key elements of AIoT systems in agriculture. In order to initiate automatic activities or offer decision assistance, these systems gather diverse data, including soil moisture, weather, crop health indicators, and equipment status. These data are then processed and analyzed. In order to improve data collection and decision-making in smart farming, this article examines new AIoT technologies. We look at the integration of various technologies, their advantages, real-world applications, issues with connectivity, security, data quality, and farmer adoption, as well as potential avenues for future research. This study uses an interdisciplinary approach to identify trends, gaps in existing practice, and tactics to optimize AIoT's influence in sustainable agriculture.

Security and Privacy in Industry 5.0: Emerging Technical Challenges and Future Pathways

Vinoth R, Omkar Singh, Navanendra Singh, Abhilasha Singh — Thu, 01 Jan 2026 00:00:00 +0000

A human-centered, resilient, and sustainable industrial ecosystem where people and intelligent systems work closely together is what Industry 5.0 offers. As industrial systems become more dispersed, data-rich, and interactive, security and privacy threats increase even as efficiency and customization gains are unlocked. The Privacy-Preserving Federated Edge Ledger (PFEL), an integrated, novel framework that combines federated learning, lightweight distributed ledgers, trusted execution environments, adaptive trust scoring, and fine-grained differential privacy to protect data and decision integrity without compromising human-in-the-loop responsiveness, is presented in this paper along with a focused analysis of the fundamental security and privacy challenges in Industry 5.0 and a survey of pertinent technical building blocks. We offer an Industry 5.0-specific threat model, describe the architecture of PFEL, outline safe model aggregation and auditability procedures, examine security and performance trade-offs, and suggest an assessment roadmap with quantifiable metrics. Lastly, we highlight future directions that harmonize security and privacy design with human-centric industrial ideals and examine wider socio-technical and legal ramifications.

AI-Powered Smart Grids: Security Challenges and Intelligent Energy Management Approaches

Abhilasha Singh, Navanendra Singh, Omkar Singh, Vinoth R — Thu, 01 Jan 2026 00:00:00 +0000

Electric power systems are being transformed into smart grids that can operate with flexibility, efficiency, and resilience thanks to the confluence of enhanced sensing, communication, distributed energy resources, and artificial intelligence. With an emphasis on security issues and clever energy management strategies, this study examines the current status of AI-powered smart grids. We present an architectural overview, pinpoint attack surfaces and threat models, and look at particular security risks such as supply-chain vulnerabilities, data integrity assaults, and privacy violations. Next, we examine AI-driven methods for distributed generation coordination, demand response, energy forecasting, and real-time optimization, and we talk about how these methods relate to privacy and security issues. We then assess defense methods such as blockchain-enabled coordination, privacy-preserving analytics, federated learning, anomaly detection, and secure communication protocols. We offer case examples and a thorough design pattern that strikes a compromise between privacy, robustness, and performance. A research agenda for safe, intelligent, and reliable smart grids, as well as suggestions for practitioners, is included in the paper's conclusion.

Adaptive Trust Models for Wireless Sensor Networks Using Blockchain and Edge Intelligence

Navanendra Singh, Abhilasha Singh, Vinoth R, Omkar Singh — Thu, 01 Jan 2026 00:00:00 +0000

Environmental sensing, smart city applications, industrial automation, and contemporary monitoring all depend on wireless sensor networks. However, there are serious trust and security issues because of their dispersed structure, resource limitations, and deployment in frequently dangerous locations. Conventional wireless sensor network trust systems rely on centralized authority and local reputation metrics, which have issues with scalability, adaptability, and resistance to complex attacks. In order to provide a reliable, scalable, and comprehensible trust management solution for wireless sensor networks, this study suggests an integrated architecture that blends adaptive trust modeling with blockchain-backed ledgering and edge intelligence. The suggested methodology records trust anchors, transaction summaries, and policy updates utilizing an immutable blockchain layer, periodic aggregation and adaptive fusion at edge devices using machine learning, and lightweight local trust estimators at sensor nodes. The system constantly modifies the trust weighting based on ambient inputs, node behavior, and context. We introduce the system architecture, formal trust update rules, a lightweight consensus and storage approach appropriate for limited contexts, and a security analysis that addresses common threats like collusion, on-off assaults, Sybil attacks, and fake data injection. When compared to baseline reputation systems, a simulation-based study shows improvements in the detection of misbehaving nodes, a decrease in false positives, and resilience against coordinated attacks. The method provides obvious routes to deployment in practical WSN applications while striking a compromise between enhanced network-level security and energy and communication overhead.

Security and Performance Assessment of Encryption Techniques for Cloud Platforms

Lakshmi Rahul Reddy Mareddy — Sat, 03 Jan 2026 00:00:00 +0000

Cloud computing platforms are becoming more popular for storage and processing of sensitive data, because of their scalability and flexibility, however the shared and distributed nature of cloud computing presents significant security dangers. Encryption is also a key element to keeping cloud data safe, but encryption operations will also add computational overhead to the system, affecting performance. This work brings an orderly evaluation of typically used encryption techniques in cloud platforms from both perspective of security effectiveness and performance efficiency. Symmetric, asymmetric and hybrid encryption approaches are examined under conditions that are relevant for cloud computing based on different criteria, e.g. encryption, decryption time, and computational overhead. Experimental results indicate that symmetric encryption performs the best on both encryption and decryption time (118 ms and 104 ms, respectively), which is applicable to bulk data protection, and asymmetric encryption causes high processing overhead (362 ms encryption time and 341 ms decryption time). Hybrid encryption offers a balanced performance where the processing times are moderate (176ms encryption and 158ms decryption) but more secure due to secure key management. The results point to some very clear tradeoffs between security strength and efficiency of performance and suggest that hybrid encryption represents a workable tradeoff for multi-user and scalable cloud environments. These findings constitute some good information in choosing encryption strategies that promote secure and efficient operations of clouds.

PERFORMANCE COMPARISON OF RELAY NODERANDOM SELECTION METHOD USING DIFFERENT SET OF PARAMETER FOR MANET

K. Thamizhmaran, A. Charles — Sat, 03 Jan 2026 00:00:00 +0000

Treads of today research uncertainty environment and natural situation if focussed in Mobile Ad hoc Networks, abig challenge to develop routing protocol that can meet different application needs and optimize routing paths according to the topology change in mobile ad hoc networks [1], [2]. The continuous transmission of small packet is called beacon packet, that advertises the presence of a base station and the mobile units sense the beacons and attempt to establish a wireless connection [3]. This research aims to propose CH-RNSR with hybrid cryptography (ECC) using RNSR algorithm. The main aim of the proposed research CH-RNSR with ECC algorithm is to increase the remaining energy with the number of malicious nodes detected during the communication via acknowledgement base than RNSR with help of one of leading simulation model called Network Simulator 2.34 work with different set of nodes, malicious nodes in same topology sizeusing various parameters such as packet delivery ratio, throughput, routing overhead, packet loss, delay and remaining energy via Network Simulator 2 (NS2).