The proliferation of large-scale Internet-of-things (IoT) deployments and the emergence of 6G wireless technologies have created a pressing need for intelligent, energy-aware, and low-latency communication frameworks. In this work, we propose a novel two-phase reinforcement learning (RL)-based architecture designed to minimize the age of information (AoI) in 6G-enabled IoT networks. Our approach integrates (i) a deep deterministic policy gradient (DDPG)-driven backscatter-assisted cognitive radio non-orthogonal multiple access (CR-NOMA) scheme in the uplink, and (ii) a lightweight Q-learning-based power-domain NOMA (PD-NOMA) strategy for the downlink. In the uplink, energy harvesting (EH) sensors employ deep RL to jointly optimize backscatter reflection coefficients and transmission scheduling over shared spectrum using CR-NOMA. This enables energy-efficient communication and reduced AoI under dynamic energy and channel conditions. In the downlink, the edge node serves multiple IoT users simultaneously using PD-NOMA, where a Q-learning agent intelligently decides whether to transmit fresh or cached data to each user based on battery levels, channel quality, and information freshness. Both phases are modeled as Markov decision processes (MDPs), allowing agents to learn independently and converge toward optimal policies that balance information freshness, spectral efficiency (SE), and energy constraints. Extensive simulations demonstrate that the proposed framework effectively reduces AoI across both phases, with consistent convergence even under varying sensor densities and EH conditions. Moreover, by relying on explainable and verifiable learning mechanisms, our model addresses emerging concerns around reliability and trustworthiness in artificial intelligence (AI)-driven 6G-IoT systems. This framework represents a step toward scalable, adaptive, and responsible AI integration for future mission-critical IoT applications.

The emergence of intent-based networking (IBN) has created new opportunities for Internet of Things (IoT) ecosystems to evolve from rigid, device-centric management toward artificial intelligence (AI)-native architectures capable of translating high-level intents into autonomous actions. In such systems, the dual requirements of information freshness and communication secrecy are critical, yet existing designs largely treat them in isolation. This paper introduces an IBN-inspired hierarchical deep reinforcement learning (HDRL) framework for reconfigurable intelligent surface (RIS)-assisted millimeter-wave (mm-wave) IoT networks threatened by unmanned aerial vehicle (UAV) eavesdroppers. The framework integrates a high-level RIS intent manager with a low-level power controller for transmit and jamming power adaptation, both trained via proximal policy optimization (PPO). A secrecy-gated transmission protocol further ensures that packets are withheld when confidentiality cannot be guaranteed. By embedding the tradeoff between age of information (AoI) and age of leaked information (AoLI) into the reward structure, the framework translates the high-level intent offresh yet securecommunication into context-aware, real-time network policies. Simulation results demonstrate significant reductions in AoI, improvements in AoLI, and enhanced secrecy efficiency under realistic fading and interference conditions. These findings underscore the potential of IBN-driven HDRL frameworks as foundational enablers for AI-powered, intent-aware, and resilient IoT communication in beyond-5G and 6G networks.

Physical layer security (PLS) has garnered increasing attention as a complementary solution to conventional crypto-graphic techniques for addressing the diversity of use cases, deployment scenarios, and device capabilities in today’s wireless networks. Integrating artificial intelligence (AI) into PLS offers a promising solution to various multi-dimensional, heterogeneous, and complex challenges stemming from the growing complexity of networks. However, the opaque nature of AI has raised concerns regarding its trustworthiness and interpretability. This paper addresses these concerns by emphasizing the crucial role of explainability in AI-based PLS in several use cases of next-generation networks. A particular focus is placed on physical layer authentication through an illustrative case study, aiming to enhance AI-empowered PLS’s practicality and trustworthiness. The paper concludes with a discussion of some critical future research directions.

Low-Earth orbit (LEO) non-terrestrial networks (NTNs) have become increasingly popular due to their ability to provide high-bandwidth communication in otherwise unserved regions. However, this comes at the cost of high satellite mobility, necessitating efficient handover (HO) strategies. Till now the focus has been on the S-band, however Ka-band is expected to be deployed for high-bandwidth applications necessitating HO studies for higher frequency bands. To address this, we evaluate the performance of the conventional received power-based HO, as well as alternatives such as elevation angle and distance-based triggers for Ka-band LEO networks, and compare them with S-band frequencies. Moreover, a dynamic threshold approach is also evaluated in terms of the number of HOs, unnecessary handovers (UHOs), radio link failures (RLFs), and distributions of downlink carrier-to-noise-plus-interference ratio (CNIR) and mean time-of-stay of a user in a cell. The obtained results indicate the efficacy of power-based mechanisms for the S-band while showing that the alternative methods provide a much better balance at the Ka-band in terms of the overhead and the overall communication link quality.

Next-generation wireless networks require massive connectivity and ubiquitous coverage, for which non-terrestrial networks (NTNs) are a promising enabler. However, NTNs, especially non-geostationary satellites bring about challenges such as increased handovers (HOs) due to the moving coverage area of the satellite on the ground. Accordingly, in this work, we compare the conventional measurement-based HO triggering mechanism with other alternatives such as distance, elevation angle, and timer-based methods in terms of the numbers of HOs, ping-pong HOs, and radio link failures. The system-level simulations, carried out in accordance with the 3GPP model, show that the measurement-based approach can outperform the other alternatives provided that appropriate values of hysteresis/offset margins and time-to-trigger parameters are used. Moreover, future directions regarding this work are also provided at the end.

Cooperative communication has been widely used to provide spatial diversity benefits for low-end user equipments, especially in ad hoc and wireless sensor networks. However, the lack of strong authentication mechanisms in these networks leaves them prone to eavesdropping relays. In this paper, we propose a secure orthogonal frequency division multiplexing (OFDM) transmission scheme, where the destination node transmits a jamming signal over the cyclic prefix (CP) duration of the received signal. Simulation results verify that as long as at least a part of the jamming signal falls to the actual data portion of the eavesdropping relay, it spreads through all the data symbols due to the fast Fourier transformation (FFT) operation, resulting in degraded interception at the eavesdropper.

Ultra-reliable low latency communication (URLLC) is undoubtedly the toughest service class in 5G-NR from the network service provider's perspective. Different methodologies have been utilized to meet the reliability and latency requirements of URLLC including proposition of various diversity techniques. This paper envisages a hybrid network using terrestrial and flying base stations to provide ubiquitous and reliable service to URLLC users. We propose utilizing the macro-diversity in the hybrid environment by exploiting the significant differences in path loss and shadowing characteristics of flying base stations (FBSs) as compared to the terrestrial base stations (TBSs). Preliminary results are presented to this effect and recommendations are made to their inclusion in the networks for supporting URLLC users in the future.