Clustered Federated Learning has emerged as an effective approach for handling heterogeneous data across clients by partitioning them into clusters with similar or identical data distributions. However, most existing methods, including the Iterative Federated Clustering Algorithm (IFCA), rely on a central server to coordinate model updates, typically requiring stable connectivity, synchronous communication rounds, and global aggregation of client models. These assumptions are difficult to satisfy in decentralized and heterogeneous environments, where clients may only have limited, local communication with a small subset of peers. As a result, such methods create a bottleneck and a single point of failure, limiting their applicability in realistic decentralized learning settings. This limitation is particularly severe in Internet of Things settings, where large numbers of resource-constrained devices, intermittent or sparse connectivity, and dynamic participation make reliance on a central server impractical. In this work, we introduce the Decentralized Federated Clustering Algorithm (DFCA), a fully decentralized clustered federated learning algorithm that enables clients to collaboratively train cluster-specific models without central coordination. DFCA uses a sequential running average to aggregate models from neighbors as updates arrive, providing a communication-efficient alternative to batch aggregation while maintaining clustering performance. Our experiments on various datasets demonstrate that DFCA outperforms other decentralized algorithms and performs comparably to centralized IFCA, even under sparse connectivity, highlighting its robustness and practicality for dynamic real-world decentralized networks.

Conventional capacitive touchscreens in mobile devices often malfunction in the presence of moisture, dust, or gloved hands, leading to unreliable user interactions. To overcome these limitations, we present TouchAI, a multi-modal touch detection and localization system that leverages built-in inertial sensors and audio interfaces. Running as a background process, TouchAI constantly monitors inertial measurements for potential touch input and then activates acoustic sensing only upon detected events, thus reducing power consumption and privacy concerns associated with continuous audio recording. We introduce a novel event identification method that distinguishes touch-start and touch-end instances, providing precise timing for data sampling. For fine-grained touch localization, TouchAI employs a lightweight Transformer-based classification model to capture spatiotemporal features from combined acoustic-inertial signals. Experiments on commercial smartphones demonstrate true positive rates of 95.3% and 99.1% for touch detection and event identification, respectively, at false positive rates only less than 3%. Ultimately, TouchAI achieves up to 97.0% and 86.5% localization accuracy on 4×2 and 6×3 touch input grids, respectively, confirming its practicality and effectiveness as an alternative interface under touchscreen malfunction scenarios.

The integration of satellite, aerial, and terrestrial networks in Satellite–Air–Ground Integrated Networks (SAGIN) enhances connectivity but also introduces new vulnerabilities to multi-layer jamming attacks. These attacks—originating from space-based, air-based, and ground-based sources—exhibit diverse signal characteristics, resource constraints, and durations, posing significant threats to communication performance and system security. A single mitigation technique is often insufficient to address these varied challenges effectively. In this paper, we propose a multi-layer adaptive jamming mitigation framework that dynamically adapts to the type of jamming encountered, with a particular focus on threats targeting Low Earth orbit (LEO) satellites within SAGIN. We evaluate a range of mitigation techniques and analyze their performance across different jamming scenarios. Our results show that tailored mitigation strategies are essential in SAGIN to achieve higher Signal-to-Noise Ratio (SNR) and lower Bit Error Rate (BER), highlighting the importance of jamming-aware defenses for enhancing the resilience and security of SAGIN systems.

For future network applications, ubiquitous connections and real-time cloud offloading are important paradigms for enabling important services. To achieve these goals, satellite networks, Free-Space Optical (FSO) communications, and Mobile Edge Computing (MEC) are key technologies. This paper proposes an efficient latency based task offloading strategy in a multi-tier Space-Air-Ground Integrated Network (SAGIN) with MEC and FSO communications. We consider a static deployment of ground users in Yamagata prefecture, Japan, offloading computational tasks to High Altitude Platforms (HAPs) and a Low Earth Orbit (LEO) satellite constellation. In this system, elevation-based FSO visibility and atmospheric attenuation can affect transmission latency, while server workload can impact computation latency. We design a hierarchical clustering-based framework and evaluate it alongside two other baseline approaches in terms of latency performance. Results show that our clustering-based task assignment achieves lower average latency and better load balancing, highlighting its potential for real-time edge-enabled FSO systems.

6G networks are expected to significantly benefit from advanced wireless local area technologies such as Wireless Gigabit (WiGig), which operates in the 60 GHz frequency band. This band supports extremely high data rates and low latency, making it ideal for next-generation wireless applications such as the metaverse and holograms. However, WiGig signals are highly susceptible to attenuation from physical obstructions, resulting in frequent handovers and connectivity disruptions. Traditional reactive handover mechanisms are often slow due to latency in decision-making and processing overhead. However, proactive handover strategies that leverage channel prediction can enhance network reliability and improve the quality of service. This paper investigates the feasibility of using statistical methods, specifically the auto-regressive integrated moving average (ARIMA) model, to predict the received signal strength indicator (RSSI) in real-world indoor WiGig environments. Our results indicate that ARIMA exhibits poor predictive accuracy, with a root mean square error (RMSE) of 15 dBm, which may trigger inaccurate handover decisions by initiating handovers under strong signal conditions or failing to respond under weak ones. To overcome this shortcoming, we propose an ensemble learning-based channel prediction approach utilizing the random forest (RF) algorithm. Our results show that the RF model significantly outperforms ARIMA by effectively capturing the nonlinear dynamics of real-world indoor WiGig channels. Specifically, the RF model achieves a 90% reduction in both mean absolute error and RMSE, and a 99% reduction in mean squared error, offering a promising solution for robust proactive handover management in 6G networks.

Emerging wireless local area networks, such as WiGig that operate in the extremely high-frequency band (60 GHz) hold significant potential for the development of next-generation 6G networks by offering high throughput and low latency. However, the 60 GHz band is prone to severe signal degradation due to channel blockages, leading to frequent handovers and challenges in maintaining seamless connectivity. Reactive handover strategies can result in service delays due to overhead and decision-making latency. To tackle these issues, proactive approaches that utilize machine learning (ML) and deep learning (DL) are becoming increasingly popular for network optimization in WiGig networks. However, existing ML/DL models are often tailored to specific network environments, making them susceptible to concept drift — a phenomenon where even minor environmental changes can significantly degrade network performance due to incorrect decision-making. This paper investigates scenarios and environmental changes that can trigger concept drift in WiGig networks. We conduct real-world experiments to analyze the statistical behavior of received signal strength, highlighting the potential for concept drift. Based on our findings, we propose a direction for identifying concept drift in WiGig networks.

Accurate traffic flow prediction is crucial for alleviating traffic congestion and optimizing intelligent transportation systems. However, traffic flow is subject to uncertainties and exhibits complex spatial and temporal dependence and dynamic change characteristics. Moreover, many efforts rely on a single view, which makes it difficult to comprehensively capture multiple levels of spatial and temporal correlations, thus limiting the accuracy of predictions. Therefore, we propose the multi-view spatio-temporal dynamic graph convolution framework MVSTDG for more comprehensively exploring and fusing the multi-view spatio-temporal features. Firstly, we design a dual-path Time-Patch Convolution (TPConv) module to separately model short-term fluctuations and long-term periodic trends, enabling effective extraction of dynamic features at multiple temporal scales. Secondly, we construct a data-driven traffic pattern library to generate dynamic adjacency matrices and integrate them with static topologies view. An Adaptive Diffusion Graph Convolutional Network (ADGCN) is then employed to model both local and global spatial correlations. In addition, we design a cross-gated spatio-temporal fusion mechanism that adaptively adjusts the contribution of short-term and long-term information, enhances the interaction of spatio-temporal information, and improves the model’s adaptive capability under different time scales. The experimental results show that MVSTDG outperforms the state-of-the-art baselines in several evaluation metrics and demonstrates higher prediction accuracy and stability on the four real datasets.

A lot of the devices in the Internet of Things are sensors responsible for capturing environmental information and relaying it to a system or network. Oftentimes, it is important to know the location of these sensors to better contextualize the information received from them. However, the sensors are purposely simple and cheap, meaning that conventional localization techniques, such as global navigation satellite systems are not feasible. Research has been done on using unmanned aerial vehicles to estimate the location of the sensors, but issues with signal strength fluctuation and location approximation because of it are still prevalent. In this work, we propose a new method for estimating the location of sensors by exploiting the characteristics of radio wave signal propagation and creating a new flight path design, a solution to minimize the impact of variation in measurements, a novel candidate point generation through signal strength analysis, and a method to find the location based on the candidates. Through a thought-out experiment, we show that the proposed algorithm is overall significantly better than the existing solution when it comes to identifying the location of outdoor sensors.

The mmWave WiGig frequency band can support high throughput and low latency emerging applications. In this context, accurate prediction of channel gain enables seamless connectivity with user mobility via proactive handover and beamforming. Machine learning techniques have been widely adopted in literature for mmWave channel prediction. However, the existing techniques assume that the indoor mmWave channel follows a stationary stochastic process. This paper demonstrates that indoor WiGig mmWave channels are non-stationary where the channel’s cumulative distribution function (CDF) changes with the user’s spatio-temporal mobility. Specifically, we show significant differences in the empirical CDF of the channel gain based on the user’s mobility stage, namely, room entering, wandering, and exiting. Thus, the dynamic WiGig mmWave indoor channel suffers from concept drift that impedes the generalization ability of deep learning-based channel prediction models. Our results demonstrate that a state-of-the-art deep learning channel prediction model based on a hybrid convolutional neural network (CNN) long-short-term memory (LSTM) recurrent neural network suffers from a deterioration in the prediction accuracy by 11–68% depending on the user’s mobility stage and the model’s training. To mitigate the negative effect of concept drift and improve the generalization ability of the channel prediction model, we develop a robust deep learning model based on an ensemble strategy. Our results show that the weight average ensemble-based model maintains a stable prediction that keeps the performance deterioration below 4%.

With the increasing demand for low-latency data processing and resource-intensive applications in vehicular networks, leveraging unmanned aerial vehicles (UAVs) for edge computing tasks has emerged as a promising solution. The lifetime of a UAV's power supply is a critical factor that can significantly impact the entire system, especially when the UAV is equipped with a server for computing user tasks. Understanding the effect of UAV failure in a cooperative scenario is crucial, particularly with regards to task latency. Thus, in this paper, we analyze the consequences of UAV failure on a cooperative UAV-enabled Multi-Access Edge Computing (MEC) system, specifically in terms of task latency. To ensure uninterrupted service delivery and maintain the quality of service, the system automatically redistributes tasks from the affected UAVs to their operational counterparts in the event of a failure. Furthermore, the paper delves into investigating the impact of severe weather condition attenuation on the millimeter wave (mmWave) and Terahertz (THz) communication channel. Specifically, it focuses on analyzing the effects on transmission time delay and energy consumption in the system. Understanding these factors is essential for optimizing mmWave and THz communication channels performance under adverse weather conditions.

This article proposes a novel algorithm, named quantum multiagent actor–critic networks (QMACN) for autonomously constructing a robust mobile access system employing multiple unmanned aerial vehicles (UAVs). In the context of facilitating collaboration among multiple UAVs, the application of multiagent reinforcement learning (MARL) techniques is regarded as a promising approach. These methods enable UAVs to learn collectively, optimizing their actions within a shared environment, ultimately leading to more efficient cooperative behavior. Furthermore, the principles of quantum computing (QC) are employed in our study to enhance the training process and inference capabilities of the UAVs involved. By leveraging the unique computational advantages of QC, our approach aims to boost the overall effectiveness of the UAV system. However, employing a QC introduces scalability challenges due to the near intermediate-scale quantum (NISQ) limitation associated with qubit usage. The proposed algorithm addresses this issue by implementing a quantum centralized critic, effectively mitigating the constraints imposed by NISQ limitations. Additionally, the advantages of the QMACN with performance improvements in terms of training speed and wireless service quality are verified via various data-intensive evaluations. Furthermore, this article validates that a noise injection scheme can be used for handling environmental uncertainties in order to realize robust mobile access.

For future networks in the 6G, it will be important to maintain a ubiquitous connection, bring processing heavy applications to remote areas, and analyze big amounts of data to efficiently provide services. To achieve such goals, the literature has utilized satellite networks to reach areas far away from the network core, and there has even been research into equipping such satellites with edge cloud servers to provide computation offloading to remote devices. However, analyzing the big data created by these devices is still a problem. One could transfer the data to a central server, but this has a high transmission cost. One could process the data through distributed machine learning, but such a technique is not as efficient as centralized learning. Thus, in this paper, we analyze the learning costs behind centralized and distributed learning and propose a hybrid solution that adaptively uses the advantages of both in a cloud server-equipped satellite network. Our proposal can identify the best learning strategy for each device based on the current scenario. Results show that the proposal is not only efficient in solving machine learning tasks, but it is also dynamic to react to different configurations while maintaining top performance.

With 6G networks, we can expect more devices to start operating in remote areas, away from the conventional network infrastructure. With an increase in the number of devices, we should also see more data that needs to be processed and analyzed. An adequate response to this scenario is using satellite networks to reach remote devices and transfer the big data generated by them to be analyzed in cloud servers through Machine Learning models. However, while this is a good solution for data analysis, it can run into bottlenecks caused by long transmission in the limited channels of satellite networks. In this paper, we will model and analyze this service model, allowing us to identify the limitations of centralized learning over satellite networks. This study manages to determine when centralized learning with remote devices is a viable solution and when it needs to be complemented by other techniques due to poor performance caused by long transmission and overloaded communication channels.

To the best of our knowledge, this article represents the first attempt toward robust beamforming design for multiuser downlink communications in intelligent reflecting surface (IRS)-enhanced satellite (SAT) and high altitude platform (HAP) integrated network. Such network configuration is mainly composed of a single-antenna SAT, a multiple-antenna HAP, multiple single-antenna SAT and HAP terminals and an IRS. Although sharing the same spectrum resource with the SAT, the HAP suspended in the near-space intends to offer temporary higher-speed and lower-delay multiuser communication connections than the SAT. Therefore, the power budget should be carefully tuned at the HAP. Toward this end, we first formally formulate the transmit power minimization problem at the HAP under the constraints of signal-to-interference-plus-noise-ratio-outage-probability (SINR-OP) at each SAT and HAP terminal by taking into account the imperfect channel state informations and their Gaussian channel estimation errors. Note that the complicated superposition of direct SAT/HAP links and cascaded IRS links, makes the optimization problem rather challenging. Specifically, we transform the probabilistic constraints into approximate deterministic ones, by performing rank relaxation. Based on this, we are able to solve the optimization problem by alternately optimizing the two ensuring subproblems. As verified by extensive simulation results, the proposed IRS-enhanced beamforming schemes can substantially diminish the transmit power at the HAP compared to the beamforming counterparts without IRS.

Multiaccess edge computing is an essential technology that academia and industry have recognized as fundamental for the future of the Internet of Things. Current research on the subject utilizes virtual machines as the intermediary between end devices and cloud servers. However, recently a new framework was proposed that utilizes Cybertwins instead of virtual machines for the same function. Such framework comes with a myriad of advantages but, most importantly, in this case, it includes a control plane capable of enabling cooperation between the Cybertwins. In this article, we present a mathematical model of the total service delay of a Cybertwin-based multiaccess edge computing system that includes user mobility, migration of virtual servers, multiple physical servers at different network tiers, fronthaul and backhaul communication, processing, and content request/caching. We also propose algorithms for guiding the operation of Cybertwins and the control plane in a multiaccess edge computing scenario. Finally, a performance analysis between Cybertwin and a virtual machine-based scheme is offered. Simulations show that Cybertwin brings significant improvement for the assumed scenario in the form of a faster overall service due to the higher cooperation. The models and simulations here were designed with the characteristics of future networks, beyond the current 5G, in mind, making them likely relevant for future networks, where multiaccess edge computing and the Internet of Things should play an even more important role.

Accurate network traffic load (TL) prediction is essential in many networking applications. However, the real TLs in practical networks may have high variability and are difficult to be predicted, which may severely affect users’ quality of experience (QoE). To address this problem, we first analyze the real-world network traffic dataset to investigate real TLs properties and find out the distance-correlation between regions in a spatial graph have the potential to improve the prediction result. Hence, we propose a time-series model based method to consider the distance-correlation in an efficient way. Empirically, experimental studies on real data demonstrate that our proposed method can effectively reduce at least 10% error value on regions with high-variability TLs. Finally, we further discuss the impact of the distance-correlation on the TL prediction.

Edge Cloud Computing is a key technology for enhancing mobile functionalities and real-time applications in devices with limited resources. This is done by sharing the resources of edge servers and offloading jobs to the edge cloud. In order to ensure a high-quality service and more efficient usage of resources, it is important not only to correctly configure the edge servers but also to carefully select where to deploy them. However, in Edge Cloud Computing there is a high amount of servers and, with the advent of 5G and Internet of Things, there will be a massive number of client devices as well. This would make the edge server deployment too complex to solve through convex techniques. In this situation, Machine Learning is the most appropriate approach. In this paper, we provide a deep analysis of the usage of k-Means Clustering and Particle Swarm Optimization in the edge cloud deployment problem. Our results show that the hyperparameters for these algorithms can significantly impact their running time as well as the efficiency of their results. Finally, we also provide how to best configure these algorithms for this specific problem.