With the emergence of next-generation communication technologies, the integration of AI into next-generation wireless networks is becoming an important step toward achieving distributed intelligence. Different distributed learning frameworks, such as federated learning (FL) and split learning (SL), enable collaborative model training by distributing computation between user equipment (UEs) and edge servers. However, FL suffers from high communication overhead due to frequent model synchronization, while SL introduces significant latency at the split layer and is sensitive to channel conditions. The effects of wireless channel dynamics on distributed learning performance and the communication-computation trade-offs across different split layers remain insufficiently studied in practical scenarios. In this work, we present a Federated Split Learning Testbed (FSLT) over 5G wireless networks that integrates FL and SL within an OpenAirInterface (OAI) software-defined radio platform. FSLT enables empirical evaluation of learning–communication trade-offs under realistic wireless conditions. The framework distributes neural network layers between UEs and edge servers, allowing investigation of channel variability, latency, and split-point selection on training performance. Experiments on an avatar skeleton extraction task demonstrate that FSLT reduces communication load by 5%-8% compared with FL and achieves faster convergence than SL under dynamic channel conditions. These results provide practical insights into optimizing distributed learning over 5G edge systems and pave the way toward AI-native 6G networks.
Nowadays, split federated learning (SFL) has emerged as an effective paradigm for enabling privacy-preserving collaborative intelligence across heterogeneous devices with limited computation. However, SFL incurs significant communication overhead in wireless networks due to the uplink transmission of high-dimensional smashed data, which degrades network efficiency. To mitigate the communication bottleneck, we propose a prototype-based SFL framework ProtoSFL. Specifically, each selected client computes local prototypes for observed classes and uploads them to the server. Based on the received prototypes, the server derives global prototypes and optimizes a weighted objective that combines classification loss with prototype alignment loss. The server then updates the model accordingly and returns personalized prototype gradients to the clients. Simulation results verify the effectiveness of ProtoSFL in reducing communication overhead, achieving a substantial reduction in uplink communication, while maintaining competitive testing accuracy under various heterogeneous data settings compared with SFL baselines.
A green quantized FSL (GQ-FSL) framework that incorporates stochastic quantization for both local collaborative training and wireless transmissions and enables large-scale DNN deployment on resource-constrained devices, achieving superior energy efficiency compared to quantized federated learning and full-precision FSL.
Federated learning (FL) has emerged as a promising paradigm for privacy-preserving edge intelligence because it enables geographically distributed devices to collaboratively train a shared model without transferring raw data to a central cloud. This capability is particularly valuable for 5G and emerging 6G networks, where edge-native services are required to satisfy stringent latency, bandwidth, and privacy constraints while operating on highly heterogeneous devices and time-varying wireless channels. In practice, however, synchronous FL is often constrained by straggling clients with limited computation capability or unfavorable communication conditions, which increases round latency and reduces overall resource efficiency. To address this challenge, this study develops a rigorously structured framework for dynamic client selection and radio resource allocation in heterogeneous wireless edge environments. Each FL round is formulated as a latency-aware scheduling problem that jointly captures local computation time, uplink transmission time, minimum participation constraints, and resource block assignment. On this basis, we propose a Dynamic Client Selection and Resource Allocation (DCS-RA) method that integrates computation-aware, channel-aware, and fairness-aware scoring with greedy resource block allocation guided by marginal completion time reduction. The study further provides a clear methodological structure, workflow visualization, literature-grounded justification, dataset documentation, and uncertainty-aware result reporting. Under the reported simulation setting with 100 clients and 20 resource blocks, DCS-RA reduces the average round completion time from 1.92 s to 1.55 s on MNIST and from 2.02 s to 1.57 s on CIFAR-10, corresponding to improvements of 19.39% and 22.47%, respectively. Standard deviation reductions of 70.59% and 80.77% further indicate improved round-to-round stability and more reliable training behavior. These results support the central conclusion that lightweight joint scheduling can materially improve wall-clock FL efficiency in heterogeneous 5G/6G edge networks.
Low Earth orbit (LEO) satellite networks are emerging as a pivotal infrastructure for global edge intelligence. In this context, integrating over-the-air (OTA) computation with adaptive beam hopping (BH) provides an innovative framework that seamlessly merges physical-layer analog aggregation with dynamic resource orchestration. This effectively overcomes the stringent bandwidth and power constraints of space platforms while extending federated learning (FL) to pervasive Internet-of-things (IoT) deployments. In this article, we first outline the fundamental principles of the dual-layer OTA model and introduce the adaptive BH mechanism designed for time-varying topologies. Then, we summarize the distinct advantages of this learning-centric architecture, which include decoupling aggregation latency from device density, optimizing spatio-temporal resource efficiency, and balancing data freshness with channel quality. Several application scenarios are explored to highlight the framework's potential across diverse vertical industries. Furthermore, a specific case is studied to demonstrate the practical efficacy of the proposed scheduling policy. The results reveal substantial performance gains in terms of model convergence speed and data utilization for satellite-based FL systems. Finally, we discuss the implementation challenges and outline future research directions, aiming to provide insights for the evolution of ubiquitous non-terrestrial intelligence.
Zhendong Li, Shaojie Wang, Zhou Su et al.· 0 citations
Open Radio Access Network (O-RAN) enables flexible and intelligent radio access network operation through disaggregation, virtualization, open interfaces, and RAN Intelligent Controllers (RICs). At the same time, the data required to train artificial intelligence and machine learning models in O-RAN is naturally distributed across user equipment, base stations, edge clouds, and management entities, which makes centralized learning costly and privacy-sensitive. Federated Learning (FL) has therefore emerged as a promising paradigm for O-RAN intelligence because it enables distributed model training without transferring raw data. In this work, we survey recent studies on the fusion of FL and O-RAN and classify them into three categories: 1) FL-assisted network control, where FL is used as a collaborative learning tool for slicing, offloading, routing, and security; 2) FL training-efficiency optimization, where communication cost, learning latency, resource consumption, and convergence are improved under O-RAN constraints; and 3) integrated approaches that jointly consider network performance and FL efficiency. Based on this taxonomy, we discuss open research challenges, including device heterogeneity, mobility, RIC integration, communication-efficient learning, and security threats, such as model poisoning and inference attacks.
Junsuk Oh, Donghyun Lee, Chunghyun Lee et al.· International Conference on...· 0 citations
The rapid growth of massive machine-type communications (mMTC), combined with advances in edge intelligence, is paving the way for low-latency, low-overhead connectivity. However, the sporadic nature of device activity in mMTC scenarios calls for efficient methods to determine which devices are active at any given time. This motivates collaborative learning within a cell-free massive multiple-input multiple-output (CF-mMIMO) architecture, where the wide geographical distribution of access points (APs) and their joint coordination make distributed learning efficient and secure. Consequently, federated learning (FL) emerges as a promising solution. Indeed, FL enables participants to train a shared model without exchanging raw local data, thereby enhancing data privacy at the AP side and lowering the fronthaul load while leveraging heterogeneous, location-dependent data. The present study proposes a novel FL framework where the CF-mMIMO participants are APs. Due to differences in device behavior, mobility patterns, and environmental factors across the network, the data collected at each AP is often non-independent and non-identically distributed (non-IID). This heterogeneity slows down the convergence of standard FL training and increase variability among client updates, particularly under heterogeneous radio feature distributions. To address this, we propose a client selection strategy that prioritizes APs based on their average received signal power. Our approach shows competitive performance compared to baseline methods, while also addressing the scalability and privacy requirements of mMTC systems. Furthermore, our study analyzes the fairness achieved by APs across devices and presents a representative, percentage-scale analysis of power-consumption gains relative to detection performance when some APs are dropped (i.e., taken out of service), examining two AP-dropping strategies. These results bring valuable insights and set guidelines towards the implementation of FL-based activity detection in CF-mMIMO networks.
Ali Elkeshawy, W. Jaafar, Haifa Fares et al.· IEEE Transactions on Machine...· 0 citations