Federated Learning via Over-the-Air Computation
Open accessThe DOI matches its Crossref record; 61 of the 62 references it cites were found in a registry or an index.
Abstract
The stringent requirements for low-latency and privacy of the emerging high-stake applications with intelligent devices such as drones and smart vehicles make the cloud computing inapplicable in these scenarios. Instead, edge machine learning becomes increasingly attractive for performing training and inference directly at network edges without sending data to a centralized data center. This stimulates a nascent field termed as federated learning for training a machine learning model on computation, storage, energy and bandwidth limited mobile devices in a distributed manner. To preserve data privacy and address the issues of unbalanced and non-IID data points across different devices, the federated averaging algorithm has been proposed for global model aggregation by computing the weighted average of locally updated model at each selected device. However, the limited communication bandwidth becomes the main bottleneck for aggregating the locally computed updates. We thus propose a novel over-the-air computation based approach for fast global model aggregation via exploring the superposition property of a wireless multiple-access channel. This is achieved by joint device selection and beamforming design, which is modeled as a sparse and low-rank optimization problem to support efficient algorithms design. To achieve this goal, we provide a difference-of-convex-functions (DC) representation for the sparse and low-rank function to enhance sparsity and accurately detect the fixe…
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“For instance, over-the-air federated learning (FL) provides a collaborative ML framework to train a global statistical model over wireless networks without accessing edge devices' private raw data [31].”
From Edge artificial intelligence for 6g: vision, enabling technologies, and applications (2022) · cites this paper as [31]
In the citing paper
“Specifically, the transmitted signals at edge devices are first multiplied by the fading channels and then superposed over-the-air with additive channel noise, resulting in a noisy weighted sum of transmitted signals [31].”
From Edge artificial intelligence for 6g: vision, enabling technologies, and applications (2022) · cites this paper as [31]
In the citing paper
“Specifically, for edge training systems, we need to jointly optimize the subcarrier and bandwidth allocation [88], [90], [241], transmit power and receive beamforming [31], [48], [178], passive beamforming at RIS [48], [178], device selection [31], [242] and activity detection [76], local updates computation [92], and global aggregation frequency control [130], thereby reducing the optimality gap”
From Edge artificial intelligence for 6g: vision, enabling technologies, and applications (2022) · cites this paper as [31]
In the citing paper
“For instance, the group sparsity can represent the combinatorial variables for edge devices selection in FL [31], edge devices activity detection [76], and inference tasks selection [53].”
From Edge artificial intelligence for 6g: vision, enabling technologies, and applications (2022) · cites this paper as [31]
In the citing paper
“Most of the resource allocation problems in edge AI need to solve a series of nonconvex optimization problems, e.g., nonconvex sparse optimization for device selection in wireless FL, nonconvex quadratic programming for transceiver design in over-the-air FL [31], lowrank matrix optimization for interference management in edge device distributed inference [52], and unit modulus constrained phase”
From Edge artificial intelligence for 6g: vision, enabling technologies, and applications (2022) · cites this paper as [31]
2 more claims about this paper
Create a free account to see every claim other papers make about this one.
Citations
Cites
Hao, X.-B., & Zhang, G.-G. (2017) · Encycl. Semantic Comput. Robotic Intell. · doi:10.1142/s2425038416300184 · Cited by 86,401 · cited 1×
“THE astounding growth in data volume promotes widespread artificial intelligent applications such as image recognition and natural language processing [1], thanks to”
BackgroundLeCun, Y., Bengio, Y., & Hinton, G. (2015) · Nature · doi:10.1038/nature14539 · Cited by 84,126
Boyd, S., & Vandenberghe, L. (2004) · doi:10.1017/cbo9780511804441 · Cited by 56,601
Vorontsova, E., Hildebrand, R., Gasnikov, A. V., & Stonyakin, F. (2010) · Computer Vision · doi:10.1007/978-1-4614-5257-7_6 · Cited by 39,270 · cited 2×
“The resulting problem is known as the sum-of-infeasibilities in the literature of optimization [39].”
Learning Multiple Layers of Features from Tiny Images
Krizhevsky, A. (2024) · doi:10.57702/zp44cu3g · Cited by 23,395
Communication-Efficient Learning of Deep Networks from Decentralized Data
McMahan, H. B., Moore, E., Ramage, D., Hampson, S., & Arcas, B. A. Y. (2016) · arXiv · doi:10.48550/arxiv.1602.05629 · Cited by 14,215 · cited 7×
“In particular, the federated averaging (FedAvg) algorithm [12] turns out to be a promising way to efficiently average the locally updated model at each device with unbalanced and non-IID data, thereby reducing the number of communication rounds between the center node and the end devices.”
InfluentialFederated Machine Learning: Concept and Applications
Yang, Q., Liu, Y., Chen, T., & Tong, Y. (2019) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/3298981 · Cited by 6,322
Han, S., Mao, H., & Dally, W. J. (2015) · arXiv · doi:10.48550/arxiv.1510.00149 · Cited by 4,662 · cited 1×
“To address this issue, there is a growing body of recent works to reduce the storage overhead, time and power consumption in the inference process using the model compression methods via hardware and software co-design [7], [8].”
MethodTowards Federated Learning at Scale: System Design
Bonawitz, K., Eichner, H., Grieskamp, W., Huba, D., Ingerman, A., Ivanov, V., Kiddon, C., Konečný, J., Mazzocchi, S., McMahan, H. B., et al. (2019) · arXiv · doi:10.48550/arxiv.1902.01046 · Cited by 2,334 · cited 2×
“It has promising applications [14], [15] such as smart retail, smart healthcare, financial services, mobile content predictions, etc.”
BackgroundFederated Learning with Non-IID Data
Zhao, Y., Li, M., Lai, L., Suda, N., Civin, D., & Chandra, V. (2018) · arXiv · doi:10.48550/arxiv.1806.00582 · Cited by 2,077 · cited 1×
“, the data is generated by distinct distributions across different devices), imposes significant statistical challenges to fit a mode from the non-IID data [13], [17].”
Background
Research works printed but not linked: 2
- rockafellar, 2015, Convex Analysis
- tao, 1997, Convex analysis approach to DC programming: Theory, algorithms and applications, Acta Math Vietn, 22, 289
Cited by
Items 901–999 of 999
Task-Driven Interference Management for Over-the-Air Computation Assisted UAV Swarms
Li, T., Leng, S.-P., Zhang, K., & Zhang, Y. (2026) · IEEE Transactions on Vehicular Technology · doi:10.1109/tvt.2026.3663097
“In AirFL systems, the impact of data aggregation quality on FL convergence has been analyzed in detail [12].”
Toward Efficient Sensing in Multi-Device ISCC by Removing Frequency Domain Redundancy
Ye, Z. Z., He, Y., Chen, W., Yu, G., & Zhang, R. (2026) · IEEE International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC) · doi:10.1109/pimrc69400.2026.11711965
“Many emerging applications further require timely processing of sensed information for intelligent decision making, which drives the evolution from ISAC to integrated sensing, communication, and computation (ISCC) [6], [7].”
Sensing-Native Over-the-Air Federated Learning
Huang, P.-Y., Gao, S.-J., Yan, J., & Giannakis, G. B. (2026) · arXiv.org · doi:10.48550/arxiv.2606.17117
“The communication-learning co-design for over-the-air FL has been extensively investigated in [5]–[10] to mitigate the impact of wireless channel fading and communication noise on the FL performance by simply assuming error-free transmission of learning statistics, i.e., means and variances of…”
Covert AirComp-Enabled Federated Edge Learning via Artificial Noise
Dong, C., Cheng, B., Ying, C., & Luo, Y. (2026) · IEEE International Symposium on Information Theory (ISIT) · doi:10.1109/isit62367.2026.11653989
“By contrast, involving more devices generally leads to improved convergence performance [3].”
InfluentialFGF-IDS: a federated framework for fine-grained intrusion detection
Yang, P., Zhang, X., Bai, F., Zeng, K., & Shen, T. (2026) · Cluster Computing · doi:10.1007/s10586-026-06193-8
uAMBIT: Byte-Budgeted Update Compression with Error-Feedback for Federated Learning
Hammami, E., Guan, P., & Taherkordi, A. (2026) · International Conference on Distributed Computing in Smart Systems and the Internet of Things (DCOSS-IoT) · doi:10.1109/dcoss-iot69657.2026.00020
“It is complementary to over-the-air aggregation [29], [30], optimizer-level drift control (e.g., SCAFFOLD/FedProx), and privacy tools (secure aggregation/DP).”
Global Unknown Estimation: A Statistical Framework for Wireless Distributed Learning
Qu, Y., Bereyhi, A., & Liang, B. (2026) · International Conference on Communications · doi:10.1109/icc59461.2026.11587058
“In federated learning, the local model parameters are computed by backpropagation over the same model architecture with the same initial point a t − 1 .”
Pang, G., & Zhang, X. (2026) · ICC 2026 - IEEE International Conference on Communications · doi:10.1109/icc59461.2026.11586876
“…this, Over-the-Air Federated Learning (Air-FL) leverages physical-layer signal superposition to aggregate models from many devices within a single time-frequency block, reducing communication delay but exposing sensitivity to channel estimation errors, interference and power con-trol [5], [6].”
Communication-Efficient Over-the-Air Federated Fine-Tuning with Heterogeneous LoRA
He, S., Zhao, Y., Shi, Y., & Zhou, Y. (2026) · International Conference on Communications · doi:10.1109/icc59461.2026.11588247
“…the estimated function variable ˆ s t is The mismatch between estimated variable ˆ s t and the actual variable s t is typically measured by the mean-squared-error (MSE), defined as (9) To simplify the design, a uniform-forcing transceiver architecture is widely adopted in existing works [15] [18].”
Self-citationOptimal Device Association for Differentially Private Over-the-Air Hierarchical Federated Learning
Qin, J.-Z., Yan, J., & Liu, H. (2026) · 2026 IEEE International Conference on Communications Workshops (ICC Workshops) · doi:10.1109/iccworkshops63917.2026.11586425
“In this paper, we assume time-invariant channels throughout the HFL process [3] [8], which allows us to analyze the properties of the optimal solution, especially the structure of optimal WD-ES association.”
Differential Privacy for Free: Wireless Federated Learning over Multiple-Access Fading Channels
Liang, H., Wen, H., Wu, K., & Xing, H. (2026) · IEEE International Conference on Communications workshops · doi:10.1109/iccworkshops63917.2026.11586391
“For multi-antenna base stations (BSs), AirFL further exploits beamforming along with device selection to enhance scalability and reduce pilot overhead [2], [4].”
Deploying Over-the-Air Federated Learning in Real-World Multi-Antenna Systems
Pradhan, S., Koç, A., Muruganandham, D., Arfaoui, M., Pietraski, P., Zhang, G.-D., Kaewell, J., & Chowdhury, K. R. (2026) · IEEE Conference on Computer Communications · doi:10.1109/infocom59046.2026.11571326
“In multi-antenna OTA-FL, prior work has primarily focused on minimizing receiver MSE, as in [20], where users’ transmit and BS’s receive beamforming are jointly optimized using alternating optimization.”
Arancioğlu, A., & Al-Hubaishi, M. (2026) · International Conference on Computational Intelligence Approaches and Applications (ICCIAA) · doi:10.1109/icciaa68481.2026.11544018
“While recent work on energy-aware FL [3] and communication-efficient aggregation [4], [5] addresses aspects of this challenge, comprehensive ablation studies quantifying the contribution of individual energy-aware components remain limited.”
Fluid Antenna-Enabled Hybrid NOMA and AirFL Networks Under Imperfect CSI and SIC
Pakravan, S., Ahmadzadeh, M., Zeng, M., Hodtani, G. A., & Li, X.-W. (2026) · doi:10.48550/arxiv.2605.11273
“To alleviate these limitations, over-the-air federated learning (AirFL) has been proposed, leveraging the superposition property of wireless multiple-access channels (MACs) to perform analog model aggregation directly over the air [6].”
Channel-Adaptive Robust Aggregation for Over-the-Air Federated Learning in Heterogeneous Networks
Fatima, Z., Shaban, Z., Jamal, Y., Shah, N., Prasad, R., & Bharath, B. N. (2026) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp55912.2026.11464541
“…equally to this work. exploiting the superposition property of the wireless multiple access channels (MAC) for simultaneous analog aggregation [5, 6], but remains highly sensitive to noise, fading, and heterogeneity, especially under straggler effects where slow clients delay or drop out of…”
Wu, C., Zheng, C., & Chen, J. (2026) · Physical Communication · doi:10.1016/j.phycom.2026.103173
Federated Learning in Distributed Computing: A Scopus-Based Bibliometric Analysis of Research Trends
Judijanto, L. (2026) · The Eastasouth Journal of Information System and Computer Science · doi:10.58812/esiscs.v3i03.993
Cai, Z., Zheng, Y., Chen, H., Liu, D., Wang, B., & Zhu, G. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2604.12401
“In OTA, all client transmits their local gradients synchronously with uncoded transmission [8], [21]–[24].”
Leveraging Kernel Symmetry for Joint Compression and Error Mitigation in Edge Model Transfer
Hamadouche, A., & Sellathurai, M. (2026) · IEEE Wireless Communications and Networking Conference · doi:10.1109/wcnc65185.2026.11555391
“Over-the-air (analog) aggregation [11] exploits the superposition property of the wireless channel to sum updates in the air, improving spectral efficiency but exposing models to channel noise and distortion.”
Adaptive Beam Hopping for Over-the-Air Online Federated Learning in LEO Satellite Networks
Wang, S.-J., Li, Z.-D., Su, Z., Zhang, Z.-H., Peng, H.-X., Cheng, N., & Chen, W. (2026) · IEEE Wireless Communications and Networking Conference · doi:10.1109/wcnc65185.2026.11555370
“Recent works such as [3], [4] demonstrated that OTA computation significantly reduces latency and spectrum usage in wireless FL, making it particularly appealing for large-scale practical networks.”
Out-of-Air Computation: Enabling Structured Function Extraction from Wireless SuperpositionPreprint
Azimi‐Abarghouyi, S. M. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2604.04312 · Cited by 6
Maduranga, M., Nandan, D., & Jayasinghe, S. (2026) · IEEE International Conference on Interdisciplinary Approaches in Technology and Management for Social Innovation (IATMSI) · doi:10.1109/iatmsi68868.2026.11465409
“Federated Learning (FL) emerged as an alternative paradigm that enables collaborative model training across distributed clients while ensuring raw data remains local [18].”
Zhang, J., Huang, J., & Huang, K. (2026) · IEEE Transactions on Communications · doi:10.1109/tcomm.2026.3717007 · Cited by 1
“In distributed sensing-and-inference systems, the scalability of multi-access is addressed by over-the-air computation (AirComp), which exploits waveform superposition to perform in-air feature aggregation and thereby enables simultaneous access [14], [15].”
Oh, J., Hong, S., Lee, D., Win, T. T., Tran, H. H., Lee, C., Song, C., Kim, G., Choi, S., & Cho, S. (2026) · International Conference on Artificial Intelligence in Information and Communication (ICAIIC) · doi:10.1109/icaiic68212.2026.11454195
“To address this, over-the-air computation (OAC) has been integrated into FL, appearing as FL over-the-air (FLOA) [5], [6].”
Event-Triggered Gossip for Distributed Learning
Zhai, Z., Yuan, X., Wang, X., Zhang, R., & Li, G. Y. (2026) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2026.3718498 · Cited by 1
“Over-the-Air (OTA) computation [28], [29], [30], [31], [32] enables simultaneous analog transmission of local gradients, such that the received waveform inherently represents their sum.”
Mitigating Heterogeneity-Induced Drift in Hierarchical Sign-Based Federated LearningPreprint
Kazemi, A., Azimi‐Abarghouyi, S. M., Fodor, G., & Fischione, C. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2602.02355
“…Internet of Things deployments with severely limited uplink rates [10], [11], wireless edge-learning systems under stringent bandwidth budgets [12], [13], and distributed sensor networks, such as in fault prediction in factories (e.g., for electrical battery manufacturing processes or…”
Distributed Perceptron under Bounded Staleness, Partial Participation, and Noisy Communication
Jain, K., Raj, A., Prakash, S., & Varma, G. (2026) · arXiv.org · doi:10.48550/arxiv.2601.10705
“In over-the-air FL, model vectors or updates are sent over broadcast or multiple-access links, and after power control, beamforming, or normalization the received vector is naturally represented as the desired model/aggregate plus an effective channel error [10], [11], [13].”
Ji, J.-C., Lam, C.-T., Wang, K., & Ng, B. (2025) · Mathematics · doi:10.3390/math14010124
A Distributed mMIMO Architecture for Over-the-Air Federated Learning
da Silva, M. V., Choi, J., Souza, R. D., & Alves, H. (2025) · IEEE Internet of Things Journal · doi:10.1109/jiot.2025.3614850
“The problems above can be addressed using over-the-air (OTA) aggregation [7], [8], which takes advantage of the superposition nature of the wireless channel to perform data aggregation during the transmission phase.”
An Intrusion Detection Framework for Fog-Enabled Smart Grid using Federated Learning
Naqi, S., & Mahmood, S. (2025) · International Conference on Intelligent Computing · doi:10.1109/icic68258.2025.11413114
Wireless Decentralized Federated Learning via Device Clustering and Inter-Cluster Link Enhancement
Zheng, W. W., Liu, H., & Zhang, Y.-J. A. (2025) · 2025 IEEE Globecom Workshops (GC Wkshps) · doi:10.1109/gcwkshps68340.2025.11591218
“Following [7], [11], the transceiver scalars that minimize the mean-squared error (MSE) of the aggregated gradients under the maximum power P 0 are given by where h ∗ mn denotes the complex conjugate of h mn .”
A heterogeneity-aware federated edge learning framework with RIS-empowered wireless aggregation
Song, T., Han, S., Wu, S., Li, H., & He, X. (2025) · International Journal of Dynamics and Control · doi:10.1007/s40435-025-01954-6
“Some studies [24, 25] have attempted to address these issues by adjusting error tolerance levels; however, these methods do not fully capture the complexities associated with performance degradation during training.”
Optimizing sparse ternary compression with thresholds for communication-efficient federated learning
Chittaiah, N. M., & Haladappa, M. S. (2025) · IAES International Journal of Artificial Intelligence · doi:10.11591/ijai.v14.i6.pp4902-4912
“However, FL encounters challenges including privacy risks, implementation difficulties, hardware constraints, communication costs, and device unavailability [3]-[5].”
Communication-Efficient Semi-Decentralized Federated Learning in the Presence of Stragglers
Li, C., Xiao, M., & Skoglund, M. (2025) · IEEE Transactions on Communications · doi:10.1109/tcomm.2025.3605479 · Cited by 3
“Remark 5: For the aggregation at the central server, over-the-air computation can be applied in A-COFFEE to integrate computation and communication seamlessly [49].”
Federated Learning in Embedded and Edge Systems: Architectures, Algorithms, and Applications
Shankar, V., Dhankher, D., Dadmi, S., Biliyar, R., & Sinchana, H. (2025) · International Conference on NexGen Networks and Cybernetics (IC2NC) · doi:10.1109/ic2nc67409.2025.11376551
“Over-the-air computation techniques [28] exploit wireless signal superposition for analog gradient aggregation, potentially reducing latency by orders of magnitude through concurrent transmission and aggregation in the physical layer.”
Machine Learning Assisted Random Access in LEO Satellite-Based Internet of Things
Woo, T.-K., Fu, C.-H., & Wu, Y. (2025) · Wireless personal communications · doi:10.1007/s11277-025-11859-4
Over-the-Air Aggregation for Min-Max Multi-Model Federated Learning over Fading Channels
Zhang, C., Dong, M., Liang, B., Afana, A., & Ahmed, Y. (2025) · Asilomar Conference on Signals, Systems and Computers · doi:10.1109/ieeeconf67917.2025.11443841
“Early OTA aggregation designs [2] are proposed for single-antenna BSs, which host the PSs. Subsequent works [3], [4] considered multi-antenna BSs and proposed various uplink beamforming schemes that are shown to significantly enhance OTA aggregation performance in wireless FL.”
Improving Efficiency of V2X Based Collaborative Perception by Over-the-Air Computation
Yu, X., Zhang, C., & Tang, S. (2025) · International Conference on Meta-Networking (MEET) · doi:10.1109/meet67398.2025.11335987
“Currently, this technology has been deployed in a variety of scenarios, including IoT [24], distributed sensor networks [25], and federated learning [26], significantly improving the timeliness of data aggregation and system resource utilization.”
Semantic Joint Source Channel Coding for Distributed Subsurface Imaging in Multi-Agent Systems
Tillmann, M. H. V., Shin, B.-S., Shutin, D., & Dekorsy, A. (2026) · IEEE Transactions on Signal and Information Processing over Networks · doi:10.1109/tsipn.2026.3719310
“On the other hand, DFC was studied in [32], with its variation with AirComp investigated in [17], [33], and deep learning based AirComp in [34], [35], [36].”
Chen, J.-J., & Liu, K.-H. S. (2025) · IEEE Vehicular Technology Conference · doi:10.1109/vtc2025-fall65116.2025.11310519
“Notice that it is also necessary to carefully select users that participate in FL to ensure a desired model accuracy [7].”
WiNPA: Wireless Neural Processing ArchitecturePreprint
Xu, S., & Du, Y. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2510.11150
“The second, “communication-empowered AI”, exploits communication networks to facilitate AI execution, such as edge computing [4], [5] and over-the-air computation [6].”
Federated-Learning Optimized AirComp Over UAV-Based CF Massive-MIMO Mobile Networks
Pang, G., & Zhang, X. (2025) · IEEE Military Communications Conference · doi:10.1109/milcom64451.2025.11310528
“Over-the-air computation (AirComp), which supports over-the-air computation-based federated learning (Air-FL), enables simultaneous transmission and aggregation of the uplink federated learning (FL) model on the CPU, reducing latency and front-haul load [3].”
Privacy Enhancement in Over-the-Air Federated Learning via Adaptive Receive Scaling
Kalarde, F. M., Liang, B., Dong, M., Ahmed, Y., & Cheng, H. T. (2026) · IEEE Conference on Computer Communications (INFOCOM) · doi:10.1109/infocom59046.2026.11571657
Robust Segmented Analog Broadcast Design to Accelerate Wireless Federated Learning
Zhang, C., Liang, B., Dong, M., Afana, A., & Ahmed, Y. (2025) · arXiv.org · doi:10.48550/arxiv.2510.02701
“Most analog schemes for wireless FL focus on the uplink aggregation of local model parameters, in either single-antenna BS [11]–[13] or multi-antenna BS [14]–[18] variants.”
Federated learning over physical channels: adaptive algorithms with near-optimal guaranteesPreprint
Zhang, R., & Mou, W. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2509.02538
“This observation motivates study of federated learning over physical communication channels [AG20b, YJSD20].”
Asy-MSFL: Communication-Efficient Federated Learning With Multiserver Adaptive Updates
Li, J., Gu, H., & Wen, H. (2025) · IEEE Internet of Things Journal · doi:10.1109/jiot.2025.3577986
“These include quantization, sparsification [7], [23] and low-rank approximations [24], [25].”
Optimizing Transceiver Design for Federated Learning Enhanced by AirComp with Byzantine Robustness
Wang, H., Chen, J., Xu, W., & Song, E.-B. (2025) · Cybersecurity and Cyberforensics Conference · doi:10.23919/ccc64809.2025.11179837
“The study on AirComp-assisted FL [5] focuses on using optimization-based methods to design the transceiver.”
On the Distribution of a Two-Dimensional Random Walk With Restricted Angles
Besser, K.-L. (2025) · IEEE Transactions on Signal Processing · doi:10.1109/tsp.2026.3694607
“The typical values of N considered in literature are around 20 to 40 [27, 28] or even larger [29].”
A Novel Heterogeneous Federated Edge Learning Framework Empowered with SWIPT
Fang, Y., Shu, S., Zhu, Y., Li, H., & Rui, K. (2025) · Symmetry · doi:10.3390/sym17071115
Zheng, J., Tian, H., Ni, W., Tian, H., & Zhang, P. (2026) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2026.3693235
“We employ a difference-of-convex-functions (DC) programming method to address the non-convexity of the rank-one constraints [40].”
Self-Adaptive Clustering Based Federated Learning Scheme in UAV Swarm Network
Li, H., Li, H., Wang, H., & Wang, P. (2025) · International Conference on Electronic Communication and Artificial Intelligence (ICECAI) · doi:10.1109/icecai66283.2025.11171063
“To accelerate global aggregation, a novel over-the-air computation based approach was proposed in [7] while achieving joint device selection and beamforming design.”
Covertness-Aware Over-the-Air Federated Learning
Ezawa, S., Chiang, Y.-H., Sun, S.-S., Lin, H. F., & Ji, Y. (2025) · IEEE Vehicular Technology Conference · doi:10.1109/vtc2025-spring65109.2025.11174591
“To facilitate FL in wireless networks, equipping FL with the over-the-air computation (denoted as OTA-FL) [2], which leverages the waveform-superposition property in multiple access channels to aggregate locally trained models wirelessly from mobile devices, has recently received significant…”
Sensing-Aware OTA-FEEL: Joint Scheduling and Beamforming Approach
Asaad, S., Wang, P., & Tabassum, H. (2025) · ICC 2025 - IEEE International Conference on Communications · doi:10.1109/icc52391.2025.11161467
“…| 2 , E 2 can be written as By using identity ( A ⊗ B ) ( C ⊗ D ) = AC ⊗ BD , and after some lines of basic derivations, we have Using the property tr { A ⊗ B } = tr { A } tr { B } , (10) reduces (9b), the aggregation error reads In this work, we follow the zero-forcing coordination strategy [10].”
MIMO Over-The-Air Federated Learning With Spiking Neural Network Via Lattice Code
Wang, C., Wu, Y., Wang, T., & Shi, Y. (2025) · International Conference on Communications · doi:10.1109/icc52391.2025.11160731
“But most of existing works, such as [5] and [6], focus on analog modulation, which may not be desirable for practical wireless communication systems that exclusively use digital modulation.”
Self-citationOver-the-Air Distributed Matrix-Vector Multiplication With Analog Coding
Choi, J. (2025) · IEEE Transactions on Vehicular Technology · doi:10.1109/tvt.2025.3533576
“This approach, known as OTA computation over multiple-access channels [19], is widely applied in federated learning [15], [25], [18].”
Transceiver Design in Over-the-Air Federated Learning over MIMO Channels
Liu, J., Chen, J., Xu, W., Cao, Z., & Song, E. (2025) · Youth Academic Annual Conference of Chinese Association of Automation (YAC) · doi:10.1109/yac66630.2025.11150237
“Prior research on AirComp FL [7], [8], predominantly optimize learning performance based on the instantaneous Mean Squared Error (MSE) of the global model, yet overlook the deterministic impact of error accumulation across multiple FL communication rounds on final training outcomes.”
Joint Graph Estimation and Signal Restoration for Robust Federated Learning
Fukuhara, T., Hara, J., Higashi, H., & Tanaka, Y. (2025) · Machine learning for signal processing ... · doi:10.1109/mlsp62443.2025.11204311
“Restrictions apply. in the global model update as a DC optimization problem and then solve it using PDCA.”
Dynamic Federated Learning Optimization: a Joint Design of Power Control and Device Scheduling
Yin, Y., Kang, W., & Liu, N. (2025) · 2025 IEEE 7th International Conference on Communications, Information System and Computer Engineering (CISCE) · doi:10.1109/cisce65916.2025.11064974
“Beamforming schemes were derived in [4], [6], [7] to minimize computation distortion.”
Second-Order Wireless Federated Leaning via Nonparametric Hessian Estimation
Hamidi, S. M., & Bereyhi, A. (2025) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp49660.2025.10889600
“GP-FL leverages analog function computation, often abbreviated as AirComp, to realize gradient aggregation directly over the air [11,22].”
InfluentialA DNN Framework on Waveform Design for Over-the-Air Computation
Evgenidis, N. G., Mitsiou, N. A., Tegos, S. A., Diamantoulakis, P. D., Sarigiannidis, P. G., & Karagiannidis, G. K. (2025) · IEEE Wireless Communications and Networking Conference · doi:10.1109/wcnc61545.2025.10978310 · Cited by 1
“Furthermore, as studied in [8], [9] OTA has been proposed as an attractive technique to enable and accelerate distributed learning techniques such as federated learning (FL), highlighting the importance of OTA computation for practical applications even in the domain of deep neural network (DNN)…”
Optimized Device Selection and Resource Management Framework for Federated Learning
Kushwaha, D., & Hegde, R. M. (2025) · National Conference on Communications · doi:10.1109/ncc63735.2025.10983624
“However, limited computational and communication resources necessitate efficient device selection and resource management to maximize participation and optimize learning performance [9], [10].”
Robust transmission design for federated learning through over-the-air computation
Abyaneh, H. Z., Asaad, S., & Rabiei, A. (2025) · China Communications · doi:10.23919/jcc.fa.2023-0282.202503
“Then, all local models are averaged at the BS to form the updated global model [3, 16, 18, 19].”
InfluentialDynamic Model Selection for Asynchronous Federated Learning in IoBT Scenarios
Tankard, E., Hagos, D. H., & Rawat, D. B. (2024) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata62323.2024.10825857
“The performance of our asynchronous FL models was evaluated and compared to several previous works: Async AirFed [47], AsyncFL [48], FedAT [49], Async-HFL [50], FedAsync [51], and AsyFL [52].”
Weighted-Sum Computation Rate Maximization for Multi-Cluster AirComp Networks
Xiao, H., Tang, X., Liang, W. J., Li, B., & Niyato, D. T. (2024) · IEEE Globecom Workshops (GC Wkshps) · doi:10.1109/gcwkshp64532.2024.11101204
“…of such a vast scale, our focus is not only the data acquired by individual wireless devices, but also the computed results of the raw data, such as the mean value of multiple temperature sensor readings, the maximum vehicle speed, or distributed model averaging in federated edge learning [2].”
UAV-Assisted Integrated Communication and AirComp with Interference Awareness
Lan, X., Tang, X., Zhang, R., Li, B., Du, Q., & Han, Z. (2024) · IEEE Globecom Workshops (GC Wkshps) · doi:10.1109/gcwkshp64532.2024.11100475
“Meanwhile, certain categories of applications (e.g., AI model updates in distributed learning [3], and temperature/humidity/chemical levels in sensing [4], [5]), require functional computations on distributed data to enable rapid collection and aggregation of large datasets for final…”
Over-the-Air Federated Learning with Model Heterogeneity: A Comparative Study
Lai, Y.-C., Chang, R. Y., & Chiu, W.-Y. (2024) · 2024 IEEE 100th Vehicular Technology Conference (VTC2024-Fall) · doi:10.1109/vtc2024-fall63153.2024.10757878
“Over-the-air FL has been proposed as a means to alleviate aggregation latency by leveraging the superposition property of wireless multiple-access channels, enabling simultaneous transmission and aggregation [11]–[14].”
Wu, M., & Chang, R. Y. (2024) · 2024 IEEE 100th Vehicular Technology Conference (VTC2024-Fall) · doi:10.1109/vtc2024-fall63153.2024.10757763
Fundamental Limits of Distributed Optimization Over Multiple Access Channel
Jha, S. K. (2024) · IEEE Transactions on Communications · doi:10.1109/tcomm.2024.3395291
“Due to its applications in federated learning [23], many interesting schemes have been recently proposed for this problem [11], [30], [32], [33], [37], [38], [39], [40].”
6G and AI: Viewpoints, Processes, and Technologies
Shoyusupov, S. A., Lanitha, B., Goyal, S., Goyal, V., Husain, S. O., & Sharma, P. (2024) · International Conference on Communication, Computing and Energy Efficient Technologies (I3CEET) · doi:10.1109/i3ceet61722.2024.10993761
Peng, N.-R., Han, Y., & Ren, L. (2024) · International Symposium on Artificial Intelligence in Medical Sciences · doi:10.1145/3706890.3706967
“Based on the distribution of participants’ data sources, federated learning is categorized into three types: horizontal federated learning, vertical federated learning, and federated transfer learning [6].”
Empowering Federated Learning with DDPG for Energy Efficiency in Communication Networks
Luu, M. N., Nguyen, M.-D., Nguyễn, V. Đ., & Pham, Q. (2024) · International Conference on Communications and Electronics (ICCE) · doi:10.1109/icce62051.2024.10634708
“Authors in [13] proposed a unique sparse and low-rank modeling strategy to enhance statistical learning performance for on-device distributed training.”
BackgroundAirFogComp: Over-the-Air-Fog Computation for Federated Learning over Fog-RAN
Park, E., & Park, S.-H. (2024) · International Conference on Ubiquitous and Future Networks · doi:10.1109/icufn61752.2024.10625104
“The process of uploading locally updated models to the CS via wireless multiple access channels (MAC) can be executed through two distinct approaches [2]: i) digital FL [3] and ii) over-the-air computation (AirComp) [4], [5].”
MethodCombining AF-based Relay and Multiantenna for Over-the-Air Computation
Tang, S., & Obana, S. (2024) · International Wireless Communications and Mobile Computing Conference (IWCMC) · doi:10.1109/iwcmc61514.2024.10592408
“The proposed relay method (Miso+CohR) is compared to AirComp with antenna selection (AntSel) [10], coherent relay with node scheduling (CohR+NS) [13], the zero-forcing (ZF) method [23], and Miso [22].”
MethodOver-the-Air Neural Group Testing
Valmeekam, C. S. K., Narayanan, K. R., & Sprintson, A. (2024) · IEEE International Conference on Machine Learning for Communication and Networking (ICMLCN) · doi:10.1109/icmlcn59089.2024.10624979
“Finally, over-the-air computation has been used in the context of federated learning in [10] and [11].”
MethodBitFool: An Optimization-Based Adversarial Attack for End-to-End Communication Systems
Yang, R., Cheng, J., Liu, D., & Li, D. (2024) · International Conference on Computer and Communication Systems (ICCCS) · doi:10.1109/icccs61882.2024.10602867
“…constraint, the objective function is maximized when It is worth noting that, to align with discussions at the physical layer of end-to-end communication, we use the Wirtinger conjugate gradient [15] instead of the commonly used gradient formed by concatenating the real and imaginary components.”
MethodOptimal Structure of Receive Beamforming for over-The-Air Computation
Zhu, H., & Qian, H. L. (2024) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp48485.2024.10446858 · Cited by 2
“…and g , which is After determining receive beamforming vector m , the optimal transmit scalars that realize the minimization of MSE are as follows [1,6] Due to the transmit power constraint, we can express η as With (3) and (4), we can further rewrite MSE as (5) Hence, we endeavor to find receive…”
BackgroundMethodZhai, Y., Huang, X., & Yu, P. (2024) · International Conference on Energy and Power Engineering, Control Engineering (EPECE) · doi:10.1109/epece63428.2024.00020
Agrawal, N., Cavalcante, R., & Stańczak, S. (2024) · International ITG Workshop on Smart Antennas · doi:10.1109/wsa61681.2024.10511748
“Problems of this type have wide-ranging applications, including, adaptive control [1], [2] and distributed learning [3].”
BackgroundOver-the-air Federated Policy Gradient
Yang, H., Huang, L., Dey, S., & Shi, L. (2024) · International Conference on Communications · doi:10.1109/icc51166.2024.10622210 · Cited by 2
“However, in many task-oriented applications, e.g., distributed learning [17]–[19], sensing [20], [21], and control [22], the exact value sent by each device is not essential, but the accumulated information of these values, such as the summation of agents’ local gradients, plays an important role.”
BackgroundHierarchical Semi-Asynchronous Federated Learning Based on Over-the-Air Computation
Jiang, J., Shi, S., Li, Y., Wang, F., & Zhang, Y. (2023) · IEEE International Conference on Advanced Infocomm Technology · doi:10.1109/icait59485.2023.10367298
“Thus, it would greatly reduce the communication resources such as bandwidth and power if clients were to perform a weighted average of the local updates while they are being uploaded [14], [15].”
BackgroundEnergy Optimization for Hierarchical Federated Learning based on Over-the-Air Computing
Zheng, K., & Li, Z. (2023) · IEEE Smart World Congress (SWC) · doi:10.1109/swc57546.2023.10448654 · Cited by 1
“Yang et al. [19] proposed using AirComp in the setting of federated learning to achieve fast global model aggregation and explore the superposition properties of wireless multi-access channels, as suggested in Figure 2.”
MethodPower Allocation and Data Assignment for Over-The-Air Distributed Learning
Guo, Y., Sung, C. W., & Shum, K. W. (2023) · International Symposium on Modeling and Optimization in Mobile, Ad Hoc, and Wireless Networks (WiOpt) · doi:10.23919/wiopt58741.2023.10349897
“Specific investigations have explored how over-the-air techniques can support distributed machine learning [9]– [13].”
BackgroundInfluentialOver-the-Air Computation Aided Federated Learning with the Aggregation of Normalized GradientPreprint
Fan, R., An, X., Zuo, S., & Han, H. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2308.09082
“In this way, the central server can get the aggregated gradient all-at-once without decoding any specific mobile device’s local gradient [6].”
BackgroundDecentralized Federated Learning With Asynchronous Parameter Sharing
Xie, H., Xia, M.-H., Wu, P., Wang, S., & Huang, K.-B. (2023) · 2023 IEEE/CIC International Conference on Communications in China (ICCC Workshops) · doi:10.1109/icccworkshops57813.2023.10233712
Model Pruning for Wireless Federated Learning with Heterogeneous Channels and Devices
Wang, D.-W., Hsieh, C.-K., Chan, K.-L., & Chien, F.-T. (2023) · Asia-Pacific Conference on Wearable Computing Systems · doi:10.1109/apwcs60142.2023.10234035
“4) Aggregation: After the base station receives all the local models, we apply the federated averaging (FedAvg) algorithm [12] in each training round.”
MethodA Momentum-Based Wireless Federated Learning Acceleration With Distributed Principle Decomposition
Dong, Y., Wang, L., Chi, Y., Hu, X., Zhang, H., Yu, F. R., & Leung, V. C. M. (2023) · IEEE International Conference on Acoustics, Speech, and Signal Processing Workshops (ICASSPW) · doi:10.1109/icasspw59220.2023.10193196 · Cited by 1
“Moreover, signal encoding methods were designed for the analog WFL systems [14, 19, 20].”
MethodSemi-Asynchronous Federated Edge Learning for Over-the-Air Computation
Kou, Z., Ji, Y., Zhong, X., & Zhang, S. (2023) · Globecom · doi:10.1109/globecom54140.2023.10437830 · Cited by 4
“To enhance the performance of wireless FEEL system, several studies related to the AirComp system for FEEL have been done [4]–[6].”
MethodMental creation and physical creation of the future of ESS
Kamabe, H. (2022) · IEICE ESS FUNDAMENTALS REVIEW · doi:10.1587/essfr.16.1_2
Analyzing Convergence Aspects of Federated Learning: More Devices or More Network Layers?
Khan, F. M. A., Hassan, S. A., Ansari, R. I., & Jung, H. (2022) · IEEE Vehicular Technology Conference · doi:10.1109/vtc2022-spring54318.2022.9860649
“The advantage provided to FL via over-the-air [14] computation has been the topic of research recently, however, the affects of wireless channel on model parameters and convergence analysis have been overlooked.”
BackgroundReconfigurable Intelligent Surfaces for Energy Efficiency in Full-duplex Communication SystemPreprint
Wang, Y., Guan, P., Yu, H., & Zhao, Y. (2022) · arXiv (Cornell University) · doi:10.48550/arxiv.2205.12079
“The non-convex rank-one constraint (19f) can be equivalently written as the following equality constraint [15]:”
BackgroundA Random Access based Approach to Communication-Efficient Distributed SGD
Choi, J. (2022) · International Conference on Communications · doi:10.1109/icc45855.2022.9838490
“As a result, for comparison, we will not consider the conventional approach, but the approach in [9], which will be discussed in Section V.”
BackgroundMethodResultInfluentialOver-the-Air Statistical Estimation of Sparse Models
Lee, C.-Z., Barnes, L. P., Zhan, W., & Özgür, A. (2021) · Globecom · doi:10.1109/globecom46510.2021.9685768
“Other authors have progressed analog over-the-air computation methods [21], [22].”
BackgroundJoint Annotator Clustering and Power Control for Energy-Efficient Wireless Crowd Labelling
Li, X., Zhu, G., Han, K., Gong, Y., & Huang, K. (2021) · IEEE Globecom Workshops (GC Wkshps) · doi:10.1109/gcwkshps52748.2021.9682025
“To acquire data for AI model training, a series of data aggregation techniques have been investigated for collecting raw data from distributed mobile devices [2]–[4].”
MethodPrivacy Amplification for Federated Learning via User Sampling and Wireless Aggregation
Seif, M., Chang, W.-T., & Tandon, R. (2021) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2021.3118408 · Cited by 60
“In [12], [13], wireless aggregation is done by aligning the gradients through power control or beamforming.”
MethodDu, H., Feng, X., Xiang, Q., & Liu, H.-Y. (2021) · arXiv.org · doi:10.48550/arxiv.2109.08819
“Given that local model transmission is an integral part of FL, there has been a growing number of studies that focus on developing novel wireless communication techniques for efficient FL (Amiri and Gündüz 2020; Yang et al. 2020).”
BackgroundMulti-task Federated Edge Learning (MtFEEL) in Wireless NetworksPreprint
Mahara, S. S., Shruti, M., Bharath, B. N., & Murthy, A. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2108.02517 · Cited by 1
Saddle Point Approximation Based Delay Analysis for Wireless Federated Learning
Yang, L., Li, L., Guo, X., Shi, Y., Wang, H., & Chen, W. (2021) · International Conference on Communications · doi:10.1109/icc42927.2021.9500298 · Cited by 2
“In [4], joint device selection and beamforming design are developed to improve the learning performance.”
BackgroundSelf-citationTowards Explainable Multi-Party Learning: A Contrastive Knowledge Sharing FrameworkPreprint
Gao, Y., Li, J., Gong, M.-G., Xie, Y., & Qin, A. K. (2021) · arXiv.org · doi:10.48550/arxiv.2104.06670
“There has been a growing interest in multi-party learning frameworks [3] [4] in the past few years.”
BackgroundSpectrum Allocation in Wireless Networks for Crowd Labelling
Li, X., Zhu, G., Shen, K., Gong, Y., & Huang, K. (2020) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp40776.2020.9053492
“One main focus in this area is on communication efficient techniques for acquiring and leveraging distributed data for AI model training [2–4].”
Method
Metadata
- Authors
- Yang, K., Jiang, T., Shi, Y., & Ding, Z.
- Affiliations
- Chinese Academy of Sciences (CN)
- ShanghaiTech University (CN)
- University of California, Davis (US)
- Venue
- IEEE Transactions on Wireless Communications
- Published
- 2020-03
- Volume
- 19
- Issue
- 3
- Pages
- 2022-2035
- ISSN
- 1536-1276 (ISSN-L)
- 1558-2248
- OpenAlex
- W2999074226
- Cited by
- 1,084
- References
- 62
- Citations by year
- 2026: 86 · 2025: 154 · 2024: 176 · 2023: 190 · 2022: 189 · 2021: 190 · 2020: 80
- FWCI
- 63.79
- Type
- journal article
- Language
- en
- Open access
- green
- arxiv.org
- Licence
- Locations
- IEEE Transactions on Wireless Communications (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Funders
- National Natural Science Foundation of China (61601290)
- National Science Foundation (CNS-1702752, ECCS-1711823)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Privacy-Preserving Technologies in Data
- Physical Sciences › Computer Science › Computer Networks and Communications › IoT and Edge/Fog Computing
- Physical Sciences › Engineering › Electrical and Electronic Engineering › Advanced Wireless Communication Technologies
- Keywords
- federated learning, over-the-air computation, edge machine learning, model aggregation, device selection, beamforming design, difference of convex functions, DC algorithm, non-IID data, data privacy
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.