In this paper, we design new reinforcement learning schemes to optimize CW sizes for the CSMA/CA protocol in IEEE 802.11 wireless technology. Each station in the wireless network is considered an ‘agent’. The states are determined by the contention window sizes, as in the binary exponential backoff algorithm, and the result of the two consecutive transmissions. The ‘action’ is considered the rule to update the contention window after each transmission. This rule is picked by the ϵ -greedy policy. In the simulations, we evaluate both the Q-learning and Double Q-learning algorithms. We apply them to the transmission history backoff probability scheme to investigate the performance of a dense network with up to 50 stations, randomly distributed in an ad hoc grid topology. The simulation results show that our schemes significantly reduce the average delay, resulting in higher throughput.
Software Defined Networking (SDN) is susceptible to attacks that can compromise system availability. To equip a network with attack countermeasures, an Intrusion Detection System (IDS) must be integrated into the SDN architecture. Such a system collects and analyses large amounts of data, including potentially sensitive and personal data, in order to respond accurately and quickly to various types of attacks. As a result, IDS design itself raises privacy concerns. Since training is performed locally on the data sources with no external party accessing raw data, Federated Learning (FL) can reduce these risks when used to train attack and anomaly detection models. This systematic literature review, conducted following the PRISMA methodology and covering IEEE Xplore, ScienceDirect, Google Schola, Wiley, and SpringerLink, identifies 24 studies published between 2019 and 2026: 16 primary studies presenting an implemented FL-based intrusion detection system for SDN and 8 prior surveys used for comparative positioning. Of 350 records initially identified, 24 met the inclusion criteria after title/abstract and full-text screening. The systematic review shows that only 2 of the 16 primary studies implement a privacy-preserving mechanism beyond keeping training data local, indicating that privacy protection in this literature remains largely nominal rather than formally guaranteed. Reported detection accuracies reach up to 99
This paper proposes a novel framework integrating Holographic Intelligent Surfaces (HIS) with magnetic energy harvesting at the source node to enable energy-efficient wireless communication. In the proposed system, energy harvesting is performed at the transmitter using magnetic field coupling, eliminating reliance on conventional power supplies. The harvested energy is then used to generate and transmit signals, which are intelligently manipulated by the HIS to enhance propagation conditions between the source and the destination. By leveraging the continuous electromagnetic control capability of HIS, the system improves signal strength, coverage, and reliability. Analytical insights highlight the potential of combining magnetic energy harvesting with HIS to support sustainable and self-powered wireless networks, particularly for next-generation communication systems.
This paper studies downlink power control in a cell-free (CF) network assisted by multiple aerial relay stations (ARSs). Under time-division duplex, each ARS applies conjugate beamforming using only uplink pilots via minimum mean-square error (MMSE) channel estimation. A closed-form use-and-then-forget achievable downlink signal-to-interference-plus-noise ratio is derived, accounting for beamforming uncertainty, noncoherent multiuser interference, and pilot-contamination-induced coherent interference. The distinctive contribution lies in a unified downlink power-control framework for cooperative multi-ARS CF networks, where max–min, max–sum, and max–product objectives are systematically investigated under the same imperfect-CSI model and per-ARS power constraints to characterize their fairness–throughput trade-offs. The max–min problem is solved by a bisection method (BS) through successive second-order cone programming feasibility checks, while a grey wolf optimizer (GWO) is developed as a unified heuristic solver for all objectives via feasible-search projection. Simulations over air-to-ground channels and deployment geometries show that increasing the number of antennas per ARS and the number of ARSs improves worst-user reliability, whereas heavier user loading and excessive pilot overhead reduce performance. The results further demonstrate that BS achieves stable convergence and consistently higher worst-user throughput than GWO for the fairness-oriented objective, while GWO provides a flexible alternative for handling multiple objectives and exposing the fairness-throughput trade-offs.
Accurate cellular traffic forecasting requires modeling both local topology-driven interactions and long-range spatio-temporal dependencies. This paper proposes ST-GAT, a dual-branch framework that combines Chebyshev graph convolution with multi-head spatial self-attention and a Temporal Transformer. An Adaptive Spatio-Topological Gating Mechanism performs input-dependent fusion of the local graph representation and the global attention representation. On the Milan Internet-traffic test set, ST-GAT with adaptive gating achieves an MAE of 77.7318, an RMSE of 136.8921, and an R^2 of 0.9683, outperforming the considered baselines under the evaluated experimental setting. Additional experiments on SMS and call traffic show that the same architecture remains competitive across different cellular service types, although the magnitude and consistency of the improvement are service-dependent. A stratified analysis of public holidays, weekends, and non-holiday weekdays further shows that the reported prediction metrics vary across calendar conditions. These results support the applicability of the proposed architecture within the evaluated offline forecasting setting, while continual adaptation, cross-city transfer, and evaluation under broader temporal conditions remain directions for future work.
