With an increasing number of devices being installed in agriculture, factories, hospitals, and cities, the need for intelligent, trustworthy, and energy-efficient communication systems is rising. This is one such system: Narrow Band Internet of Things, or NB-IoT, which connects devices at a long range with low-power consumption. But it also has a poor reaction time and is not adaptable to changing situations. This review discusses how artificial intelligence (AI), including machine learning (ML), deep learning, and other intelligent methods, are making NB-IoT more adaptable and predictive. By deploying lightweight AI models and edge computing—that is, processing locally on a device—systems are able to better manage wireless signals, save battery life, and improve overall performance. It also discusses several related real-world applications in the area of agriculture, healthcare, industry, and smart cities and achieves benefits such as saving water, early disease diagnosis, predictive machine maintenance, and environmental monitoring. The integration of AI with NB-IoT fosters digital solutions toward sustainability, security, and being human-centric. All the progress made thus far will be of immense utility to decision-makers, engineers, planners of smart cities, and service providers who look to smarter, more affordable, privacy-friendly technology for the future.
Artificial intelligence (AI) and machine learning (ML) are increasingly being applied to antenna modeling, performance prediction, optimization, inverse design, and reconfiguration for Internet of Things (IoT) applications. This article presents a systematic literature review and evidence mapping to identify AI approaches used in antenna design and optimization and examine their performance, validation, and IoT-related requirements. Studies published between January 2021 and June 2025 were identified through IEEE Xplore, ScienceDirect, Scopus, SpringerLink, Wiley Online Library, and MDPI, using predefined search and eligibility criteria. A total of 35 reports, comprising 23 primary AI antenna studies and 12 review or survey articles, were included and classified by AI approach, antenna design task, application domain, validation level, and IoT requirements. The literature shows that supervised learning, deep learning, evolutionary and metaheuristic optimization, surrogate modeling, and hybrid approaches are used across antenna development. AI supports performance prediction, design space exploration, parameter optimization, and reduction of selected EM evaluations within specific workflows. However, reported accuracy, computational efficiency, and antenna performance improvements vary with datasets, antenna configurations, validation procedures, and computational methods, limiting direct quantitative comparison. Experimental and measured studies provide stronger evidence of practical feasibility than simulation-only or conceptual investigations, while AI-assisted methods generally complement rather than replace physics-based EM simulation and final verification. Key challenges include dataset standardization, generalization to unseen configurations, inconsistent reporting, resource constraints, and limited experimental evidence for adaptive AI-controlled antenna systems. Future research should emphasize reproducible datasets, leakage-resistant evaluation, uncertainty-aware modeling, quantitative AI–EM comparisons, resource-efficient implementation, and experimental validation.
Symbiotic wireless communication (SWC) is a novel paradigm, which allows several networks to coexist and share resources with each other through mutual interactions, just as it is the case within the natural ecological systems. The approach has been perceived as a viable measure of resolving energy inefficiency and scarcity of the spectrum in the future wireless communication networks. Nevertheless, sharing of resources and services in SWC places has not been an easy task because of untrustworthy trust systems across diverse platforms. To deal with this challenge, this paper will present a symbiotic blockchain wireless network (SBWN) which will combine blockchain technology and SWC to provide safe and effective communication. The framework proposed makes use of cognitive backscatter communication and energy-efficient sharding mechanism to minimize the processing overhead and improve system performance. The simulation results show that SBWN ensures up to 60
Semantic communication is expected to become a foundational technology in 6G edge intelligence networks, enabling end devices to transmit task-relevant semantic features rather than symbol-level data. However, semantic streams are highly time-correlated and often reveal latent identity or contextual information, making them vulnerable to inference attacks when processed by semi-trusted edge servers. Moreover, existing privacy-preserving mechanisms lack the ability to provide stateless, fine-grained temporal authorization while maintaining the ultra-low latency required by 6G applications. This work introduces T–PAC, a temporal-policy access control framework designed specifically for secure and privacy-preserving semantic streaming in 6G edge environments. T–PAC employs (i) time-indexed key derivation with embedded key transitions, enabling stateless per-timestamp decryption through lightweight authorization tokens; (ii) a semantic desensitization pipeline combining redaction, pseudonymization, generalization, and differential privacy to reduce semantic leakage; and (iii) a policy mapping mechanism that supports dynamic, window-bounded access control based on temporal requirements. Extensive experiments show that T–PAC achieves sub-microsecond encryption latency on desktop platforms and microsecond-level latency on resource-constrained edge devices, while maintaining compatibility with semantic feature processing. These results highlight the suitability of T–PAC as a secure foundation for next-generation 6G semantic communication systems.
This paper presents a multi-task spatiotemporal deep learning framework designed for zero-shot geographic generalization in predictive channel modeling for low Earth orbit (LEO) satellite communications at Q-band (39 GHz). Unlike many existing methods that rely on location-specific training, the coordinate-based long short-term memory (LSTM) architecture processes historical sequences of atmospheric and geographic data spanning 60 min to simultaneously predict weather conditions and excess path loss (EPL) for 5 h, incorporating inherent Gaussian uncertainty quantification. The proposed framework is designed to improve geographic generalization, which remains a challenge for existing machine learning approaches in satellite communications. A comprehensive evaluation conducted across various European climates demonstrates exceptional dual performance, achieving a root mean square error (RMSE) of 0.026 dB at the training locations and an average error of 0.213 dB across ten entirely unseen European cities, demonstrating strong generalization and representing a substantial improvement over existing methods. The framework maintains consistent prediction performance across all untrained European cities, while the uncertainty estimation mechanism provides calibrated confidence measures that may support operational decision-making. Furthermore, the coordinate-based inference strategy enables prediction at previously unseen locations without location-specific retraining, making the proposed framework a promising approach for large-scale predictive link adaptation in future LEO satellite networks. Although the proposed framework achieved promising results on ITU-R model-generated datasets, further validation using independent real-world satellite link measurements is needed to confirm its practical applicability under operational conditions.
