Federated Learning (FL) enables collaborative model training across distributed clients while preserving data privacy, yet its robustness under fine-grained data perturbations remains insufficiently characterised. Pixel-level manipulations, such as the One Pixel Attack, are known to degrade model performance in centralized settings, but their behaviour under federated aggregation is less understood. This paper investigates the optimisation-level impact of pixel-level perturbations in FL through a diagnostic framework termed Federated One Pixel Perturbation. By systematically varying perturbation intensity, the number of compromised clients, data heterogeneity, and aggregation rules, we examine how localised input disturbances propagate through aggregation and influence global training dynamics. To contextualise these effects, the proposed perturbation is compared with random pixel perturbation, label flipping, FGSM, and PGD baselines. Experiments on CIFAR-10, CIFAR-100, and FEMNIST reveal a consistent divergence between final accuracy and convergence behaviour. Under moderate pixel-level perturbations, FL can maintain relatively stable final accuracy while requiring substantially more communication rounds. In 100-client non-IID settings, update divergence, gradient variance, and cosine similarity further show that structured perturbations alter client-update dynamics even when accuracy changes remain limited. Adversarial training improves final accuracy, but does not fully restore stable update behaviour. Overall, structured pixel-level perturbations primarily disrupt optimisation dynamics before visible accuracy degradation occurs. The results reveal a threshold-like robustness boundary and highlight the limitation of accuracy-centric robustness evaluation in federated learning.

With the rapid expansion of internet-of-things (IoT) networks, ensuring efficient data collection has become a key challenge, especially in remote and hard-to-reach areas. Unmanned aerial vehicles (UAVs) offer a flexible solution for IoT networks. However, UAV communications face challenges such as signal attenuation, interference, and line-of-sight (LoS) constraints. To address these limitations, reconfigurable intelligent surfaces (RISs) are proposed as a promising solution to enhance UAV communications. In this paper, we propose a novel federated learning (FL)-based framework for optimising RIS-assisted UAV networks. Our approach integrates deep reinforcement learning (DRL) for UAV trajectory planning and IoT device scheduling while leveraging FL to train UAV models collaboratively without sharing raw data. Additionally, a block coordinate descent (BCD) algorithm is employed to optimise RIS phase shifts. This framework enhances communication reliability, energy efficiency, and scalability, making UAV-assisted IoT networks more adaptive to dynamic environments.

Light-emitting-diode based visible light positioning (VLP) system(s) can provide high-precision positioning accuracy, but require an unobstructed line-of-sight (LoS) environment. Visible light multipath signals are significantly affected by temporal and spatial dispersion, making them less effective for stringent positioning services. This work addresses the LoS limitation by optimizing multipath reflections and creating a sustainable alternative channel, even in a fully blocked LoS scenario. Accordingly, optical intelligent reflecting surfaces (IRS) are proposed to converge incident light toward a desired direction with an optimal energy threshold. Unlike recent multi-step IRS orientation and mismatched alignment techniques for position estimation, this paper aims to improve the positioning accuracy of a deterministic (fixed but unknown) receiver by modeling three types of reflections (i.e., fully diffuse, partially diffuse, and purely directive). In addition, a maximum likelihood estimation technique is proposed to leverage the simplicity and robustness of the received signal strength trilateration and direct positioning techniques within a synchronous indoor VLP system. Our simulation results demonstrate 88% improvement in positioning accuracy for partially diffused reflections over LoS-only case.

Predicting signal blockages in millimetre-wave and terahertz networks is essential for enabling proactive handover (PHO) and ensuring seamless connectivity. Existing approaches utilising deep learning, multi-modal vision and wireless sensing data primarily depend on centralised model training. Although these techniques are effective, they come with high communication costs, inefficient bandwidth usage, and latency issues, which restrict their real-time applicability. This paper proposes a Semantic-Aware Federated Blockage Prediction (SFBP) framework, leveraging the lightweight computer vision technique MobileNetV3 for edge-based semantic extraction, lowering communication and computation costs. Furthermore, we introduce a Similarity-Driven Federated Averaging (SD-FedAVG) mechanism to enhance the robustness of the model aggregation process, effectively mitigating the impact of noisy updates and adversarial attacks. Our proposed SFBP framework achieves 97.1% blockage prediction accuracy, closely matching centralised learning methods, while reducing communication costs by 88.75% compared to centralised learning and by 57.87% compared to FL without semantic extraction. Moreover, on-device inference reduces the latency by 23% compared to centralised learning and 18% compared to FL without semantic extraction, improving real-time decision-making for PHO. Additionally, the SD-FedAVG mechanism improves prediction accuracy under noisy conditions, directly impacting the PHO by reducing the handover failure rate by 7%.

Emergency communication is critical but challenging after natural disasters when the ground infrastructure is devastated. Unmanned aerial vehicles (UAVs) have enormous potential for agile relief coordination in such scenarios. However, effectively leveraging UAV fleets poses additional challenges, in terms of security, privacy, and efficient collaboration across response agencies. This paper presents a robust blockchain-enabled framework to address these challenges by integrating a consortium blockchain model, smart contracts, and crypto-graphic techniques to securely coordinate UAV fleets for dis-aster response. Specifically, we make two key contributions: a consortium blockchain architecture for secure and private multi-agency coordination and an optimized consensus protocol balancing efficiency and fault tolerance using a delegated proof of stake practical Byzantine fault tolerance (DPoS-PBFT). Com-prehensive simulations show the framework's ability to enhance transparency, automation, scalability, and cyber-attack resilience for UAV coordination in post-disaster networks.

Millimeter wave (mmWave) bands come with a deployment challenge of signal degradation when an obstacle blocks the line of sight (LOS) link. This paper proposes an intra-cell proactive handover (PHO) framework utilizing links from an intelligent reflective surface (IRS) to assist a 60GHz mmWave transmitter. Empowered by vision-aided wireless communication (VAWC) the PHO is aiming to replace a blocked LOS link with an IRS-assisted link. Evaluation of IRS-assisted link performance during the HO scenario is conducted to establish a solid understanding of deployment requirements and limitations. Estimations of the received signal strength indicator (RSSI) were performed to compare IRS-assisted links in the blocked area and LOS links in the absence of a blockage event. Results showed that for an IRS to provide a comparable signal level to the original LOS link, beam focusing must be the operating mode. IRS-assisted PHO scenarios were evaluated based on a range of IRS elements (64, 100 and 1000) to compare between the two links. Signal drop was between 30 to 15 dBm depending on the number of IRS elements and user location. The gap in signal level was further reduced to 10–5 dBm by increasing the number of antenna elements in the uniform linear array (ULA) sector transmitting to the IRS. Finally, the results showed that a HO to a 30GHZ IRS-assisted link with 64 ULA antenna elements and 1000 IRS elements will perform comparably to the LOS signal strength.

The millimetre-wave (mmWave) based frequency-modulated continuous wave (FMCW) radar is a pioneering technique for non-invasive vital sign monitoring, specifically designed for integration within 5G/Beyond 5G (B5G) network environments. In this study, we focus on evaluating the accuracy of the FMCW radar in measuring heart rate (HR) and respiratory rate (RR) at various heights relative to the subject's chest. To harness the low-latency benefits of advanced wireless technologies, the mmWave radar operates in tandem with the real-time data-capture adapter (DCA1000) evaluation module, facilitating the high-speed transmission of vital sign data. We employed two signal-processing methodologies, fast fourier transform (FFT) and peak count, to analyze and validate the radar's precision and consistency against reference sensors. Our results underscore the effectiveness of the peak count method, which demonstrates superior accuracy, with mean absolute error (MAE) values of 5.33 breaths/min for RR and 4.03 beats/min for HR, along with root mean square error (RMSE) values of 5.30 breaths/min for RR and 4.30 beats/min for HR, consistent across all tested heights. The proposed system significantly enhances the reliability and responsiveness of next-generation healthcare systems, offering a robust solution for patient monitoring in various interactive and intelligent real-time settings.

This article introduces a new method to improve the dependability of millimeter-wave (mmWave) and terahertz (THz) network services in dynamic outdoor environments. In these settings, line-of-sight (LoS) connections are easily interrupted by moving obstacles like humans and vehicles. The proposed approach, coined as Radar-aided Dynamic blockage Recognition (RaDaR), leverages radar measurements and federated learning (FL) to train a dual-output neural network (NN) model capable of simultaneously predicting blockage status and time. This enables determining the optimal point for proactive handover (PHO) or beam switching, thereby reducing the latency introduced by 5G new radio procedures and ensuring high quality of experience (QoE). The framework employs radar sensors to monitor and track object movement, generating range-angle and range-velocity maps that are useful for scene analysis and predictions. Moreover, FL provides additional benefits such as privacy protection, scalability, and knowledge sharing. The framework is assessed using an extensive real-world dataset comprising mmWave channel information and radar data. The evaluation results show that RaDaR substantially enhances network reliability, achieving an average success rate of 94% for PHO compared to existing reactive HO procedures that lack proactive blockage prediction. Additionally, RaDaR maintains a superior QoE by ensuring sustained high throughput levels and minimising PHO latency.

To accommodate the growing number of heterogeneous radios with limited wireless resources, symbiotic communication (SC) inspired by biology has been recently proposed to establish a symbiotic radio (SR) ecosystem. In this SR ecosystem, through collaboratively optimizing service/resource exchange policies, radios can coevolve like organisms, thus enabling various radio resources (such as spectrum, energy, and computing power) to complement each other. However, one critical challenge is securing a trusted coevolution environment in an SR ecosystem since the SRs with different network operators should coevolve under unreliable wireless links with complex electromagnetic interference. Moreover, multidimensional resources participated and a wide array of service requirements pose additional challenges to service/resource exchange decision-making across massive SRs. In this paper, we propose a Blockchain-empowered Intelligent cOevolution scheme for SRs, named BIO-SR. Specifically, BIO-SR exploits the digital acyclic graph (DAG) blockchain consensus in securing a trusted environment of SRs and applies deep reinforcement learning (DRL) in service exchange decision-making. The simulation results show that the BIO-SR scheme outperforms conventional solutions in terms of transmission rate and latency under both non-attack and malicious attack scenarios.

Unmanned aerial vehicles (UAVs) and reconfigurable intelligent surfaces (RISs) are expected to be widely used in future wireless communication networks to improve spectrum and energy efficiency. In this paper, RIS-assisted UAV communication systems are studied and analysed by developing a comprehensive mathematical framework for examining their outage performance. In order to study the effect of the RIS on UAV communications, two system scenarios are considered: in the first scenario, the UAV acts as an aerial base station (BS) serving a ground user to offload the terrestrial network, and in the second system, the UAV acts as an aerial user served by a terrestrial BS. We present channel models considering the UAV’s unique characteristics, propose a closed-form approximation for the signal-to-noise-ratio (SNR) distribution, and derive an analytical expression for the relevant outage probability. Results show that RIS can significantly improve the performance of UAV communication systems by introducing energy-efficient and reliable links. This opens the door for UAV networks, which are highly scalable, adaptable, and robust to environmental changes. Furthermore, the results show that the UAV position and altitude optimisation significantly affects the outage performance.

Line of sight (LoS) links that use high frequencies are sensitive to blockages, making it challenging to scale future ultra-dense networks (UDN) that capitalise on millimetre wave (mmWave) and potentially terahertz (THz) networks. This paper embraces two novelties; Firstly, it combines machine learning (ML) and computer vision (CV) to enhance the reliability and latency of next-generation wireless networks through proactive identification of blockage scenarios and triggering proactive handover (PHO). Secondly, this study adopts federated learning (FL) to perform decentralised model training so that data privacy is protected, and channel resources are conserved. Our vision-aided PHO framework localises users using object detection and localisation (ODL) algorithm that feeds a multiple-output neural network (NN) model to predict possible blockages. This involves analysing images captured from the video cameras co-located with the base stations (BSs) in conjunction with wireless parameters to predict future blockages and subsequently trigger PHO. Simulation results show that our approach performs remarkably well in highly dynamic multi-user environments where vehicles move at different speeds, and achieves 93.6% successful PHO. Furthermore, the proposed framework outperforms the reactive-HO methods by a factor of 3.3 in terms of latency while maintaining a high quality of experience (QoE) for the users.

Smart energy performance monitoring and optimisation at the supplier and consumer levels is essential to realising smart cities. In order to implement a more sustainable energy management plan, it is crucial to conduct a better energy forecast. The next-generation smart meters can also be used to measure, record, and report energy consumption data, which can be used to train machine learning (ML) models for predicting energy needs. However, sharing energy consumption information to perform centralised learning may compromise data privacy and make it vulnerable to misuse, in addition to incurring high transmission overhead on communication resources. This study addresses these issues by utilising federated learning (FL), an emerging technique that performs ML model training at the user/substation level, where data resides. We introduce FedraTrees, a new, lightweight FL framework that benefits from the outstanding features of ensemble learning. Furthermore, we developed a delta-based FL stopping algorithm to monitor FL training and stop it when it does not need to continue. The simulation results demonstrate that FedraTrees outperforms the most popular federated averaging (FedAvg) framework and the baseline Persistence model for providing accurate energy forecasting patterns while taking only 2% of the computation time and 13% of the communication rounds compared to FedAvg, saving considerable amounts of computation and communication resources.

Abstract Using reconfigurable intelligent surfaces (RISs) to improve the coverage and the data rate of future wireless networks is a viable option. These surfaces are constituted of a significant number of passive and nearly passive components that interact with incident signals in a smart way, such as by reflecting them, to increase the wireless system's performance as a result of which the notion of a smart radio environment comes to fruition. In this survey, a study review of RIS‐assisted wireless communication is supplied starting with the principles of RIS which include the hardware architecture, the control mechanisms, and the discussions of previously held views about the channel model and pathloss; then the performance analysis considering different performance parameters, analytical approaches and metrics are presented to describe the RIS‐assisted wireless network performance improvements. Despite its enormous promise, RIS confronts new hurdles in integrating into wireless networks efficiently due to its passive nature. Consequently, the channel estimation for, both full and nearly passive RIS and the RIS deployments are compared under various wireless communication models and for single and multi‐users. Lastly, the challenges and potential future study areas for the RIS aided wireless communication systems are proposed.

The upsurge in wireless devices and real-time service demands force the move to a higher frequency spectrum. Millimetre-wave (mmWave) and terahertz (THz) bands combined with the beamforming technology offer significant performance enhancements for future wireless networks. Unfortunately, shrinking cell coverage and severe penetration loss experienced at higher spectrum render mobility management a critical issue in high-frequency wireless networks, especially optimizing beam blockages and frequent handover (HO). Mobility management challenges have become prevalent in city centres and urban areas. To address this, we propose a novel mechanism driven by exploiting wireless signals and on-road surveillance systems to intelligently predict possible blockages in advance and perform timely HO. This paper employs computer vision (CV) to determine obstacles and users’ location and speed. In addition,this study introduces a new HO event, called block event (BLK), defined by the presence of a blocking object and a user moving towards the blocked area. Moreover, the multivariate regression technique predicts the remaining time until the user reaches the blocked area, hence determining best HO decision. Compared to conventional wireless networks without blockage prediction, simulation results show that our BLK detection and proactive HO algorithm achieves 40% improvement in maintaining user connectivity and the required quality of experience (QoE).

The spectrum efficiency can be greatly enhanced by the deployment of machine-to-machine (M2M) communications through cellular networks. Existing resource allocation approaches allocate maximum resource blocks (RBs) for cellular user equipments (CUEs). However, M2M user equipments (MUEs) share the same frequency among themselves within the same tier. This results in generating co-tier interference, which may deteriorate the MUE’s quality-of-service (QoS). To tackle this problem and improve the user experience, in this paper, we propose a novel resource utilization policy, which exploits reinforcement learning (RL) algorithm considering the pointer network (PN). In particular, we design an optimization problem that determines the optimal frequency and power allocation needed to maximize the achievable rate performance of all M2M pairs and CUEs in the network subject to the co-tier interference and QoS constraints. The proposed scheme enables the user equipment (UE) to autonomously select an available channel and optimal power to maximize the network capacity and spectrum efficiency while minimizing co-tier interference. Moreover, the proposed scheme is compared with traditional spectrum allocation schemes. Simulation results demonstrate the superiority of the proposed scheme than that of the traditional schemes. Moreover, the convergence of the proposed scheme is investigated which reduces the computational complexity (CC).

In this paper, we quantify the effect of residual hardware impairments (RHI) on the error rate performance of a relay-based non-orthogonal multiple access (NOMA) system, where the communication between a source node and multiple users is completed via an amplify-and-forward (AF) relay node. In particular, we focus on the pairwise error probability (PEP) analysis and derive an accurate PEP approximation to characterize the performance of NOMA users under Rayleigh fading channels. The derived PEP expression is then exploited to investigate the diversity gain and the union bound on the bit error rate (BER) of the underlying system. Our results demonstrate that the presence of RHI causes an error floor at high signal-to-noise ratio (SNR) values. This error floor yields a detrimental effect on the achievable diversity order of NOMA users, where it is shown that the diversity order of all users converges to zero.

In this paper, we investigate the application of differential modulation in simultaneous wireless information and power transfer (SWIPT) relay networks. Considering time switching (TS) and power splitting (PS) receiver architectures, we adopt a moments-based approach to derive novel expressions for the outage probability and throughput of SWIPT relay systems with the amplify-and-forward (AF) relaying protocol. We quantify the impact of several system parameters involving the energy conversion efficiency and the TS and PS ratio assumptions, imposed on the energy harvesting (EH) relay terminal. Our results reveal that the throughput performance of the TS protocol is superior to that of the PS protocol at lower receive signal-to-noise (SNR) values, which is in contrast to point-to-point SWIPT systems. A Monte Carlo simulation study is presented to corroborate the proposed analysis.

In this letter, the performance of differential modulation in simultaneous wireless information and power transfer (SWIPT) cooperative amplify-and-forward (AF) networks is investigated. In particular, we derive novel closed-form expressions for the probability density function (pdf) of the end-to-end signal-to-noise ratio (SNR) and the average bit error rate (ABER) of the considered SWIPT cooperative scenario. Based on the derived results, we analyze the impact of the underlying system parameters on the system performance. Numerical results show that the optimum location of the relay terminal is closer to the source than to the destination. Moreover, it is demonstrated that the value of the power splitting (PS) ratio at the relay significantly impacts the system performance. The results of Monte Carlo simulations are provided to corroborate the analysis.

The present work is devoted to the analytic performance evaluation of cooperative spectrum sensing (CSS) over generalized fading channels. The proposed analysis is based on the efficient Gaussian-Finite-Mixture (GFM) that allows the derivation of a simple and accurate closed-form expression for the average probability of energy detection (ED) under different fading environments. Capitalizing on this, we derive generalized closed-form expressions for the global probabilities of detection for the CSS with two main hard centralized fusion rules, namely, the AND and the OR rules. The efficiency and usefulness of the proposed expressions is justified by comparing the corresponding complementary receiver operating characteristic (ROC) curves for both multipath and composite multipath/shadowing fading channels, which are otherwise particularly difficult to obtain. The offered analytic results are corroborated by respective results from computer simulations and it is shown that the corresponding performance depends significantly on both the severity of fading and the involved number of users in the collaborative network.