The DOI matches its Crossref record; 61 of the 62 references it cites were found in a registry or an index.
Federated Learning via Over-the-Air Computation
Open accessAbstract
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 801–900 of 999
Resource-Efficient and Delay-Aware Federated Learning Design under Edge Heterogeneity
Nickel, D. R., Lin, F., Hosseinalipour, S., Michelusi, N., & Brinton, C. G. (2021) · 2022 IEEE International Conference on Communications Workshops (ICC Workshops) · doi:10.1109/iccworkshops53468.2022.9814610 · Cited by 2
VoiceMoji: A Novel On-Device Pipeline for Seamless Emoji Insertion in Dictation
Kumar, S., Harichandana, B. S. S., & Arora, H. (2021) · IEEE India Conference · doi:10.1109/indicon52576.2021.9691564 · Cited by 3
“While this helps in user personalization and prediction with out-of-vocabulary (OOV) words, it presents its challenges of network availability, security breaches as showcased by [17].”
BackgroundOver-the-Air Federated Learning with Retransmissions (Extended Version)Preprint
Hellström, H., Fodor, V., & Fischione, C. (2021) · arXiv.org · doi:10.48550/arxiv.2111.10267 · Cited by 2
On the Necessity of Aligning Gradients for Wireless Federated Learning
Chang, W.-T., Seif, M., & Tandon, R. (2021) · IEEE International Workshop on Signal Processing Advances in Wireless Communications (SPAWC) · doi:10.1109/spawc51858.2021.9593177 · Cited by 2
“…been proposed: a truncation-based strategy was proposed in [3] for improving the signal-to-noise ratio (SNR) for unbiased gradient aggregation, [4], [11] proposed to jointly design users’ transmit powers and a denoising factor at the PS to minimize the mean-squared-error (MSE) between the…”
BackgroundReconfigurable Intelligent Surface Assisted Federated Learning with Privacy Guarantee
Yang, Y., Zhou, Y., Wang, T., & Shi, Y. (2021) · IEEE International Conference on Communications workshops · doi:10.1109/iccworkshops50388.2021.9473728 · Cited by 4
“In AirComp, multiple edge devices transmit data concurrently via the same frequency channel and exploit the waveform superposition property of multiple-access channel to achieve model aggregation [8], [9].”
InfluentialSelf-citationStart Making Sense: Semantic Plane Filtering and Control for Post-5G ConnectivityPreprint
Popovski, P., & Simeone, O. (2019) · arXiv.org · doi:10.48550/arxiv.1901.06337 · Cited by 2
Yang, H., Wang, Z.-K., & Cui, L.-Z. (2026) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2026.3668371 · Cited by 1
“Research [37] shows that selecting more clients per round can improve FL performance.”
A review of federated learning: architectures, challenges, and targeted solutions
Velaei, M., Ghahramani, H., Ghaffari, A., & Derakhshanfard, N. (2026) · Cluster Computing · doi:10.1007/s10586-026-06286-4 · Cited by 1
“In the AirComp method in FL, the superposition property of signals in the wireless multi-user channel is used to calculate the weighted average of local model updates directly and simultaneously on the server; instead of each device sending its local model individually [ 212 , 213 ].”
Distributed Edge Learning under Imperfect Data Sensing
Karbalayghareh, M., Love, D. J., & Brinton, C. G. (2026) · arXiv.org · doi:10.48550/arxiv.2607.18649 · Cited by 1
“Substantial work has analyzed FL convergence under practical impairments, including non-i.i.d. data, partial participation, imperfect channel conditions, and over-the-air aggregation distortion [8]–[11], but most analyses treat training data as a fixed noise-free input.”
Fast Pinching-Antenna Activation for AirComp
Cheng, Z.-Q., Zhao, B.-Q., Ouyang, C.-J., & Zhang, X.-Q. (2026) · doi:10.48550/arxiv.2607.08873 · Cited by 1
“As a result, the end-to-end channel from user k to the PA activated at waveguide m is Let Following the classical AirComp model [3], the BS aims to recover the desired function, namely the sum through the waveform superposition property of the wireless multiple-access channel, where s = [ s 1 , . .…”
Lan, X., Tang, X., Zhang, R.-N., Du, Q.-H., & Quek, T. Q. S. (2026) · IEEE Journal on Selected Topics in Signal Processing · doi:10.1109/jstsp.2026.3705715 · Cited by 1
“…and high spectral efficiency, AirComp is particularly attractive for delay-critical applications that compute aggregate functions over a large amount of data instead of reconstructing individual data, such as consensus control, collaborative sensing, and distributed machine learning [12], [13].”
OFLight: Lightweight Gradient Compression for Over-the-Air Federated Learning
Zhu, J. F., Yang, H. Q., Παππάς, Ν., & Poor, H. V. (2026) · IEEE International Conference on Sensing, Communication, and Networking (SECON) · doi:10.1109/secon68281.2026.11579004 · Cited by 1
Photonics-aided THz over-the-air computation: a proof-of-concept experiment
Che, M., Chen, H., Yano, T., & Kato, K. (2026) · Optical Review · doi:10.1007/s10043-026-01067-1
“[8–12].”
FedMECA: Scalable Federated Learning via Memory-Efficient and Concurrent Aggregation
Chen, Z., Zhang, D., & Lu, X. (2026) · ACM Conference on AI and Agentic Systems (CAIS) · doi:10.1145/3786335.3813142 · Cited by 1
Resource-Element Energy Difference for Noncoherent Over-the-Air Federated LearningPreprint
Chen, H., & Bozorgasl, Z. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2605.07263 · Cited by 2
“A large body of OTA-FL work has been developed under a coherent AirComp model, in which analog update symbols are phase-and amplitude-aligned at the receiver to compute a sum or weighted sum [1]–[7].”
Ahmadzadeh, M., Pakravan, S., Ajib, W., Zeng, M., Hodtani, G. A., & Wang, J. (2026) · IEEE Internet of Things Journal · doi:10.1109/jiot.2026.3701371 · Cited by 1
“Compared to conventional orthogonal transmission schemes, overthe-air federated learning (OTA-FL) can dramatically reduce communication latency and bandwidth consumption, making it a promising enabler of scalable edge intelligence [5], [6], [7].”
Zheng, P., Zhu, Y., Yuan, X.-P., Hu, Y.-L., & Schmeink, A. (2026) · IEEE Transactions on Wireless Communications · doi:10.48550/arxiv.2604.26741 · Cited by 1
“…are robust to small perturbations of model parameters, analog transmission leveraging over-the-air computation (AirComp) has emerged as an effective technique to address the communication bottleneck of FL, yielding what is known as over-the-air federated edge learning (AirFEEL) [7], [8], [9].”
Pakravan, S., Ahmadzadeh, M., Zeng, M., Hodtani, G. A., Li, X., Wang, J., & Wang, G. (2026) · IEEE Transactions on Vehicular Technology · doi:10.1109/tvt.2026.3686744 · Cited by 1
“This approach not only streamlines the aggregation process but also optimizes the use of available communication resources, contributing to a more efficient and scalable FL framework [5], [6], [7].”
Energy-Aware Bid-Based Client Selection for Federated Learning in Resource-Constrained IoT Networks
Albelaihi, R. (2026) · The 1st International Conference on Emerging Tech & Innovation (ICETI) · doi:10.3390/cmsf2026013007 · Cited by 1
“Energy-aware participation, like threshold rules, restricts training to clients with sufficient residual energy [8,12].”
Chen, Y.-L., Liang, H., & Tu, X.-M. (2026) · IEEE International Conference on Acoustics, Speech, and Signal Processing · doi:10.1109/icassp55912.2026.11463544 · Cited by 1
“A prominent approach is over-the-air FL (AirFL) [2], which leverages the superposition property of radio-frequency waveforms to allow simultaneous transmission and efficient model aggregation.”
Asaad, S., Tabassum, H., & Wang, P. (2026) · arXiv.org · doi:10.48550/arxiv.2603.15783 · Cited by 1
“As the figure shows, CollabSenseFed tracks both cases closely, indicating a practically-negligible impact of target sensing on learning quality [31], [36].”
MD-AirComp+: Adaptive Quantization for Blind Massive Digital Over-the-Air Computation
Qiao, L., Wang, Y., Jiang, H., Liu, X., Xing, Y., Wu, Y., Gao, Z., & Zhou, X. (2026) · Chinese Journal of Electronics · doi:10.23919/cje.2025.00.447
“The use of multi-input multi-output (MIMO) systems for AirComp is explored in [18].”
Intelligent O-RAN Optimization: AI/ML-Enabled Dynamic Prediction for Adaptive Rate Control
Kumar, S. (2026) · Journal of Network and Systems Management · doi:10.1007/s10922-025-10016-3 · Cited by 1
Timely Parameter Updating in Over-the-Air Federated Learning
Zhu, J., Zhao, Z.-Y., Li, X., Du, R., Jin, S., & Yang, H. (2025) · arXiv.org · doi:10.48550/arxiv.2512.19103 · Cited by 1
Mitigating Jamming Attacks in Over-the-Air Federated Learning via Coordinated Dropout
Ezawa, S., Nishimoto, K., Chiang, Y.-H., Sun, S.-S., Chiang, T.-W., Lin, H., & Ji, Y. (2025) · Globecom · doi:10.1109/globecom59602.2025.11432204 · Cited by 1
“To this end, equipping FL with the over-the-air computation (denoted as OTA-FL) [4]–[6], which leverages the waveform-superposition property in multiple access channels to aggregate locally trained models wirelessly from AIoT devices, has recently received growing research attention.”
Optimized Multi-Receiver OTA Data Fusion: A Distributed Resource Control Approach
Charatsaris, P., Siraj, M. S., Diamanti, M., Tsiropoulou, E., & Papavassiliou, S. (2025) · Conference on Network Function Virtualization and Software Defined Network · doi:10.1109/nfv-sdn66355.2025.11349531 · Cited by 1
“Another line of work focuses on improving the efficiency of OTA-based data fusion through intelligent selection of edge devices , considering the tradeoff between aggregation accuracy and resource constraints [10].”
Zhang, X.-L., Deng, Y.-S., & Mahmoodi, T. (2025) · IEEE Transactions on Vehicular Technology · doi:10.1109/tvt.2026.3707692 · Cited by 1
“2) In Synchronous (Sync) FL, the global aggregation is triggered only when updates from all users are uploaded [7], [8].”
ECPRESSO: Efficient Gradient Compression for Over-the-Air Federated Learning
Zhu, J., Du, R., Zhao, Z., Quek, T. Q. S., & Yang, H. H. (2025) · International Workshop on Radio Frequency (RF) Computing · doi:10.1145/3737906.3767101 · Cited by 1
Efficient Multi-Worker Selection Based Distributed Swarm Learning via Analog Aggregation
Yao, Z.-Y., Wang, Y., Zhang, S.-Y., Li, Y.-S., Cai, Z.-P., & Tian, Z. (2025) · Asilomar Conference on Signals, Systems and Computers · doi:10.1109/ieeeconf67917.2025.11443871 · Cited by 1
“Some communications theory-based designs benefits in mitigating the complexity of communication scheme with source multiplexing prototypes, An insightful solution termed over-the-air (OTA) aggregation performs as the reliable channel modeling and more efficient channel multiplexing for multi-access…”
Employing Federated Learning in Satellite Communication Networks for Global Connectivity
Kumbhar, V. R., Panchami, S. V., Jadhav, V. D., Poongodi, S. K., Menaria, S., & Shendage, J. D. (2025) · International Conference on Intelligent Communication Networks and Computational Techniques (ICICNCT) · doi:10.1109/icicnct66124.2025.11232878 · Cited by 1
“This approach is particularly beneficial for adaptive traffic signal control in smart cities, where traffic dynamics can vary significantly across intersections, times of day, and seasonal patterns [5][6].”
Dynamic Cell Association for Hierarchical Over-the-Air Federated Learning with Non-IID Data
Zhen, Z., Zhou, F., Qiu, X., & Hong, S. (2025) · International Conference on Computer Communications and Networks (ICCCN) · doi:10.1109/icccn65249.2025.11133763 · Cited by 1
“Based on the basic theory of AirComp [10], in order to achieve integrated uplink communication and model aggregation between UEs and BSs, we design the following composite function. where ψ k ( · ) = 1 is the pre-processing scalar at the k -th UE and φ l ( · ) = 1 / |K l | is the post-processing…”
Integrated User Scheduling and Beam Steering in Over-the-Air Federated Learning for Mobile IoT
Liu, S., Fu, N., Zhang, Z., Huang, Y., & Quek, T. Q. S. (2025) · ACM Transactions on Internet Technology · doi:10.1145/3747348
“Although the centralized gradient decent method can be applied to update the model w in order to obtain w , it becomes impractical as the local data set D 𝑘 model based on the local dataset D 𝑘 where h 𝑘 𝑘 𝑘 , m is the aggregate beam steering vector, n is the additive white Gaussian noise…”
CF-mMIMO-Based Computational Offloading for UAV Swarms: System Design and Experimental Results
Sun, J., Lin, H., Shi, W., Xu, W., & Wang, D.-M. (2025) · Electronics · doi:10.3390/electronics14132708 · Cited by 1
Stochastic Device Scheduling and Power Control in Federated Learning with Energy Harvesting
Zhang, B. H., Dallal, E., & Lu, N. (2025) · International Conference on Communications · doi:10.1109/icc52391.2025.11161225 · Cited by 1
“Likewise, an adaptive computation and power control scheme is introduced in [15], enabling devices to optimize their transmission power and fully utilize computational and communication resources.”
Weighted Over-the-Air Federated Learning
Azimi‐Abarghouyi, S. M., Tassiulas, L., & Fischione, C. (2025) · IEEE International Conference on Machine Learning for Communication and Networking (ICMLCN) · doi:10.1109/icmlcn64995.2025.11140426 · Cited by 1
“Another strategy involves joint device selection and power allocation schemes [12]–[15].”
Optimizing Power for Interference-Resilient Hierarchical Over-the-Air Federated Learning
Yao, J., He, D., Wen, J., Fang, Y., Xu, J., & Han, G. (2025) · International Conference on Electronics, Circuits and Information Engineering (ECIE) · doi:10.1109/ecie65947.2025.11087072 · Cited by 1
“In single-cell scenarios, various strategies have been proposed to mitigate these issues to improve aggregation accuracy and system reliability, such as gradient compression [3], adaptive power control [4], device scheduling [5], and beamforming [6].”
Over-the-Air Inference over Multi-hop MIMO Networks
Bian, C.-H., Hua, M., & Gündüz, D. (2025) · arXiv.org · doi:10.48550/arxiv.2505.00430 · Cited by 1
“For instance, linear operations such as summations and polynomials can be carried out by exploring the signal-superposition property of the wireless medium, which have been widely applied in federated learning [3], [4].”
Unified Robust and Convergent Over-The-Air Federated Learning under Channel Noise and HeterogeneityPreprint
Shaban, Z., Shah, N., & Prasad, R. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2503.19549 · Cited by 1
“Over-The-Air Federated Learning (OTA-FL) strategy becomes a preferred choice for efficient communication over a common uplink Multiple Access Channel (MAC) [9]–[11].”
Quantized Analog Beamforming Enabled Multi-task Federated Learning Over-the-air
Yao, J., Xu, W., Zhu, G., Yang, Z., Huang, K., & Niyato, D. T. (2025) · IEEE Vehicular Technology Conference · doi:10.1109/vtc2025-spring65109.2025.11174840 · Cited by 1
“The potential multiplexing capability of multi-antenna systems to support multi-task AirComp has rarely been considered in recent works.”
Power-Efficient Over-the-Air Aggregation With Receive Beamforming for Federated Learning
Kalarde, F. M., Dong, M., Liang, B., Ahmed, Y., & Cheng, H. T. (2025) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2025.3626720 · Cited by 1
Zhou, H., Chiang, Y.-H., Chen, C., Wang, X.-Y., & Ji, Y.-S. (2025) · Consumer Communications and Networking Conference · doi:10.1109/ccnc54725.2025.10975912 · Cited by 1
“Several studies have been devoted to showing the effectiveness of OTA-FL (Over-the-Air Federated Learning) to tackle these issues [2], [4], [5].”
Channel-Estimation-Free Gradient Aggregation for OFDM Over-the-Air Federated Learning
Zeng, Z., Zhong, C., & Yuan, X. (2024) · International Conference on Wireless Communications and Signal Processing (WCSP) · doi:10.1109/wcsp62071.2024.10826990 · Cited by 1
“Prior efforts have addressed channel fading through techniques like channel inversion on the transmitter side [1]–[3], and joint optimization of beamforming vectors at both transmitters and receivers [4], [5].”
A Survey on Device Scheduling in Over-the-Air Computation-Enabled Federated Learning
Iqbal, U., Jung, H., & Shin, H. (2024) · International Conference on Information and Communication Technology Convergence (ICTC) · doi:10.1109/ictc62082.2024.10827320 · Cited by 1
“The combination of AirComp with FL enables the latter to enhance its communication efficiency and reduce the latency energy consumption, as AirComp can allow the server to send the local updates simultaneously [3].”
Wu, S., Shen, S., Yeoh, P. L., & Lim, T. J. (2024) · IEEE Annual Congress on Artificial Intelligence of Things (AIoT) · doi:10.1109/aiot63253.2024.00011 · Cited by 1
“Most such methods exclude some clients or model parameters in each aggregation step, resulting in a negative impact on accuracy and convergence speed [11] [12].”
BackgroundComputation Selection: Scheduling Users to Enable Over-the-Air Federated Learning
Nazer, B., & Narayanan, K. R. (2024) · IEEE International Symposium on Information Theory (ISIT) · doi:10.1109/isit57864.2024.10619217 · Cited by 1
Adaptive Communication Resource Allocation for Federated Learning with UEP Strategies
Lan, M., Xiao, S., & Zhang, W. (2024) · IEEE Vehicular Technology Conference · doi:10.1109/vtc2024-spring62846.2024.10683286 · Cited by 1
“Some studies focus on uncoded transmission methods like over-the-air computation [3], [4], digressing from digital communication systems.”
BackgroundAirComp-Based Federated Learning Enhanced by IRS for Secure Transmission in IIoT
Han, H., Wang, H., Zhai, W., Zhang, Y., Huang, G., Feng, Y., Zhang, Q., & Lu, W. (2024) · International Conference on Electronics, Circuits and Information Engineering (ECIE) · doi:10.1109/ecie61885.2024.10627140 · Cited by 1
“As a non-orthogonal multiple access (NOMA) technology, over-the-air computation enables resource-efficient and low-latency analog FL by utilizing the waveform superposition property of multiple-access channels [5].”
MethodByzantine-Resilient Hierarchical Federated Learning with Clustered Over-The-Air Aggregation
Nordlund, D., Liao, J., & Chen, Z. (2024) · IEEE International Conference on Acoustics, Speech, and Signal Processing Workshops (ICASSPW) · doi:10.1109/icasspw62465.2024.10626690 · Cited by 1
Interference Cancellation for OTFS-Based Over-the-Air Computation
Huang, X., Hellström, H., & Fischione, C. (2024) · 2024 IEEE International Conference on Communications Workshops (ICC Workshops) · doi:10.1109/iccworkshops59551.2024.10615413 · Cited by 1
“Most previous studies, e.g., [2], [3], considered AirComp over time-invariant channels due to the simpler mathematical analysis.”
BackgroundRobust Over-the-Air Federated Learning
Kim, H., Nam, H., & Love, D. (2024) · Annual Conference on Information Sciences and Systems · doi:10.1109/ciss59072.2024.10480195 · Cited by 1
“However, the aggregation of local updates on edge devices becomes a pivotal role, and the uplink rate limitations remain a notable bottleneck when employing orthogonal transmission [3].”
BackgroundJi, J., Lam, C., Wang, K., & Ng, B. K. (2023) · IEEE International Conference on Computer and Communications (ICCC) · doi:10.1109/iccc59590.2023.10507265 · Cited by 1
“This system aims to minimize the global loss function F ( w ) [24]. where w is a vector.”
BackgroundWang, Z.-W., Huang, K., & Eldar, Y. (2023) · Global Communications Conference · doi:10.1109/globecom54140.2023.10437548 · Cited by 1
“AirFL is enhanced by beamforming [9], gradient pruning [10], and broadband transmission [11].”
BackgroundIRS-Assisted Digital Over-the-Air Federated Learning
Pan, Y., Wang, Z., Wu, L., & Zhou, Y. (2023) · Globecom · doi:10.1109/globecom54140.2023.10436811 · Cited by 1
Resource Efficient Over-the-Air Fronthaul Signaling for Uplink Cell-Free Massive MIMO Systems
Shaik, Z. H., Thoota, S. S., Björnson, E., & Larsson, E. G. (2023) · ICC 2024 - IEEE International Conference on Communications · doi:10.1109/icc51166.2024.10622362 · Cited by 1
Nguyen, K. A., Nguyen, Q., & Hong, J. (2023) · Sensors · doi:10.3390/s23167184 · Cited by 1
“FL adopts a distributed training approach, where each edge device trains a common model with its own local data samples in a distributed manner and forwards the locally trained model to the parameter server for subsequent operations [5].”
BackgroundXiao, S., & Wu, C.-Y. (2023) · Future Generation Computer Systems · doi:10.1016/j.future.2023.06.027 · Cited by 4
AirComp-aided Safety-aware CAM Broadcast Rate Control in C-V2X Sidelink
Hsieh, D.-Y., Kuo, J.-J., Chen, W.-T., & Sheu, J. (2023) · IEEE Vehicular Technology Conference · doi:10.1109/vtc2023-spring57618.2023.10199437 · Cited by 1
“It has been widely used for federated learning in the network edge [13] and in the wireless network control system [14], [15].”
MethodFederated Edge Learning via Integrated Sensing, Computation, and Communication
Liu, P., Zhu, G., Wang, S., Wen, M.-W., Luo, W., Poor, H. V., & Cui, S. (2023) · ICC 2023 - IEEE International Conference on Communications · doi:10.1109/icc45041.2023.10279085 · Cited by 1
“While interesting, the works in [4]–[8] all took the learning error as the objective function.”
BackgroundMethodFundamental Limits of Distributed Optimization over Multiple Access Channel
Jha, S. K., & Mayekar, P. (2023) · IEEE Information Theory Workshop (ITW) · doi:10.1109/itw55543.2023.10161622 · Cited by 1
Energy Efficient Wireless Crowd Labeling: Joint Annotator Clustering and Power Control
Li, X.-Y., Zhu, G., Shen, K., Han, K., Huang, K.-B., & Gong, Y. (2023) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2022.3208198 · Cited by 1
“To acquire data for FEEL, a series of wireless data aggregation techniques have been investigated for transmitting data from distributed mobile devices to the server [16], [17], [18], [19], [20].”
MethodQuick and Reliable LoRa Physical-layer Data Aggregation through Multi-Packet Reception
You, L.-Z., Tang, Z., Wang, P., Wang, Z.-R., Dai, H.-P., & Fu, L. (2022) · arXiv.org · doi:10.48550/arxiv.2212.06582 · Cited by 1
“[31]–[33] considers OFDM-based AirComp for federated edge learning.”
Liu, Z.-R., Li, Z., Wen, M.-W., Gong, Y., & Wu, Y.-C. (2022) · arXiv.org · doi:10.48550/arxiv.2212.00697 · Cited by 1
How to Coordinate Edge Devices for Over-the-Air Federated Learning?Preprint
Sedaghat, M. A., Bereyhi, A., Asaad, S., & Mueller, R. R. (2022) · arXiv (Cornell University) · doi:10.48550/arxiv.2211.03449 · Cited by 2
“2This is in particular effective, since the MSE of estimated global model directly affects the learning quality of the FL scheme [17].”
BackgroundMethodInfluentialHassani, M., & Gholizadeh, R. (2022) · arXiv.org · doi:10.48550/arxiv.2210.16917 · Cited by 1
“This method has two major benefits, namely (i) accessing to larger data provided by different clients, and (ii) utilizing the computing power of all the clients to train a general model [1], [2], [3], [4], [5], [6], [7], [6], [8].”
MethodAnalog Over-the-Air Federated Learning With Real-World Data
Chen, Z., Li, Z., & Xu, J. (2022) · IEEE International Conference on Sensing, Communication, and Networking (SECON) · doi:10.1109/seconworkshops56311.2022.9926339 · Cited by 1
“To address the mentioned communication bottleneck, the analog over-the-air computation technique is proposed [5], [6].”
BackgroundMethodInfluentialDynamic Clustering and Power Control for Two-Tier Over-the-Air Federated Learning
Guo, W. M., Huang, C. Z., Qin, X., Yang, L. X., & Zhang, W. (2022) · IEEE/CIC International Conference on Communications in China (ICCC) · doi:10.1109/iccc55456.2022.9880666 · Cited by 3
“However, the distortion caused by the additive noise and channel fading of the wireless channels degrades the FL performance, and existing works to address this issue exhibit similar solutions that they discard the devices with relatively weak channel gains [5], [6], which results in insufficient training data exploit…”
BackgroundFundamental Limits of Over-the-Air Optimization: Are Analog Schemes Optimal?
Jha, S. K., Mayekar, P., & Tyagi, H. (2022) · IEEE Journal on Selected Areas in Information Theory · doi:10.1109/jsait.2022.3177582 · Cited by 8
“Similar scaling approaches are also presented in [19], [21]–[23].”
BackgroundResultEnabling Efficient Scheduling Policy in Intelligent Reflecting Surface Aided Federated Learning
Wang, P., Li, L., Wang, D., Ma, D., & Han, Z. (2021) · Globecom · doi:10.1109/globecom46510.2021.9685096 · Cited by 1
“Besides, over-the-air computation (AirComp) is a promising data aggregation method in wireless communication, which can achieve faster convergence speed and improve the global model [5].”
MethodInfluentialResearch on Transmission and Offloading Scheme of MEC-IRS for Distribution Network Service
Xia, B., & Tang, Y.-C. (2021) · Journal of Physics, Conference Series · doi:10.1088/1742-6596/2087/1/012074 · Cited by 1
“Literature [6-7] uses the edge calculation assisted by intelligent reflective surface to directly analyze the data of sensors and other equipment, and then screen the information threatening the stable operation of distribution network business, and then upload it to the cloud for processing.”
BackgroundSpatio-Temporal Federated Learning for Massive Wireless Edge Networks
Liu, C., Feng, K., Wei, L., & Luo, Y. D. (2022) · International Conference on Communications · doi:10.1109/icc45855.2022.9838401 · Cited by 1
An Overview of Enabling Federated Learning over Wireless Networks
Foukalas, F. T., Tziouvaras, A., & Tsiftsis, T. A. (2021) · IEEE International Mediterranean Conference on Communications and Networking (MeditCom) · doi:10.1109/meditcom49071.2021.9647687 · Cited by 1
“Further, sparsification [3] has been proposed, which reduces the FL communication cost by transmitting only a few parameters of a model [19], instead of broadcasting the whole model.”
BackgroundUnit-Modulus Wireless Federated Learning Via Penalty Alternating Minimization
Wang, S., Li, D., Wang, R., Hao, Q., Wu, Y., & Ng, D. W. K. (2021) · Globecom · doi:10.1109/globecom46510.2021.9685788 · Cited by 1
Communication-Efficient Distributed SGD using Preamble-based Random AccessPreprint
Choi, J.-H. (2021) · arXiv.org · doi:10.48550/arxiv.2105.09427 · Cited by 1
OFDM-Assisted Over-the-Air Computation With Privacy Protection in Federated Learning
Zuo, S., Fan, R., Zhao, P., Zhan, C., Hu, H., & Xiong, Z. (2026) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2026.3715532
“Regarding high-speedtransmissioninMIMOtechnology,[21]introduces asparseandlow-rankmodelingapproachtoenhancealgorithm efficiencyusingthedifference-of-convexmethod.”
FedMaJ: Ordinal Meta-Aggregation for Federated Learning Using Majority-Judgment Voting
Choukri, D., Sabir, E., Elbiaze, H., & Haqiq, A. (2026) · doi:10.48550/arxiv.2610.39985
“In over-the-air FL (OTA-FL), this choice also interacts with transmit scaling, receiver denoising, and aggregation distortion [6], [7].”
Sarker, M., & Cherkaoui, S. (2026) · doi:10.48550/arxiv.2610.37424
“(Corresponding author’s e-mail: manobendu.sarker@polymtl.ca) FL [2]–[4].”
A Multi-Step Inertial ADMM Algorithm for Federated LearningPreprint
Li, H., Wang, Q., Yan, X., Yan, C., & Han, D. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2610.24536
“Owing to its privacy-preserving nature, FL has attracted considerable attention and has been widely applied in various areas such as mobile edge computing [17, 31, 33, 36], the Internet of Things [9], digital health [27], and vehicular networks [5, 25, 26, 28].”
Learning OTA: A Unified Framework for Edge Sensing, Computation, and CommunicationPreprint
Karbalayghareh, M., Love, D. J., & Brinton, C. G. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2610.04718
Liang, Y.-P., Chen, Q.-M., Wen, D.-Z., Zhu, G.-X., Awan, M. K., Zhao, M.-M., & Jiang, H. (2026) · IEEE Transactions on Vehicular Technology · doi:10.1109/tvt.2026.3701541
“To enhance communication efficiency, over-the-air federated learning (OTA-FL) has emerged as a promising solution [13], [14], [15].”
Over-the-Air Federated Learning in Heterogeneous Mobile Wireless Networks
Xiang, M., Michelusi, N., Eldar, Y. C., & Su, L. (2026) · ACM International Symposium on Theory, Algorithmic Foundations, and Protocol Design for Mobile Networks and Mobile Computing (MobiHoc) · doi:10.1145/3842721.3849688
Zhang, L., Jia, Y., Cao, L., Wang, D., Ren, J., & Cui, S. R. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2610.03741
“AirComp has also been widely used for FL. Over-the-air FL aggregation was proposed to reduce communication latency during distributed training [14].”
Jang, J., Lyu, H., & Yang, H. J. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2609.23084
“We consider a single-cell network in which K EDs are served by a PS, co-located with the access point [5], [8].”
InfluentialDistributed CNN inference system based on aircomp
Liu, Q., Du, X., Tian, C., Hu, H., & Wang, J. (2026) · Wireless Networks · doi:10.1007/s11276-026-04216-1
“In the edge intelligent network, [16, 17] propose the AirComp-assisted FL, which enables the server to directly obtain the weighted-sum local models during uplink aggregation, achieving a lower communication delay.”
Occupancy-Domain Over-the-Air ComputationPreprint
Azimi‐Abarghouyi, S. M. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2609.11289 · Cited by 1
“Receiver-side designs move the burden to the server through denoising-factor optimization, beamforming and device selection [3], [4], and The author is with the Department of Electrical Engineering, Chalmers University of Technology, Gothenburg, Sweden (e-mail: azimimo@chalmers.se). blind designs…”
Non-Coherent Over-the-Air Federated Learning: Protocol, Convergence, and Device Scheduling
Wen, H.-F., Michelusi, N., Simeone, O., Yang, Y., & Xing, H. (2026) · doi:10.48550/arxiv.2609.08312
“The MIMO extension in [7] jointly designed device selection and beamforming to reduce aggregation error.”
Ziaeddini, A., Yakimenka, Y., & Kliewer, J. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2609.03245
“When multiple nodes transmit simultaneously, the i -th receiver directly obtains an analog sum of the signals [41]–[43], where h ji is the wireless channel coefficient, P j is the transmit power, and n i,t is Gaussian noise.”
Li, Z., Wang, S., Su, Z., Zhang, Z., Peng, H., Cheng, N., Wang, Y., & Chen, W. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2609.03202
“To address the communication-efficient aggregation challenge, over-the-air (OTA) computation, also known as Air-Comp, has been advocated as a physical-layer technique that exploits the waveform superposition property of wireless multiple-access channels [15].”
Yang, P., Yang, L., Kuang, W., & Wang, S. (2026) · IEEE Internet of Things Journal · doi:10.1109/jiot.2026.3709331
“Recently, AirComp has gained significant popularity as a key enabler for over-the-air federated learning (AirFL), resolving the communication bottleneck in large-scale distributed learning networks [5], [6], [7].”
Li, Z.-D., Wang, S.-J., Su, Z., Luan, T., Sun, R., Wang, Y., & Chen, W. (2026) · doi:10.48550/arxiv.2608.13885
“This approach exploits the superposition property of wireless channels to aggregate analog signals simultaneously, significantly enhancing spectral efficiency [7].”
InfluentialImplementation and Validation of an Air Quality Prediction Model Based on Federated Learning
Xu, J.-X. (2026) · Mathematical Modeling and Algorithm Application · doi:10.54097/7bq52c49
“This method not only protects the confidentiality of data, but also benefits from the knowledge of multiple data sources [15, 16].”
Wireless Linear Computation Broadcast
Tan, S., & Jafar, S. A. (2026) · IEEE Transactions on Communications · doi:10.1109/tcomm.2026.3742805
“The coupling of communication and computation has been predominantly explored in uplink multiple-access channel (MAC) settings, where waveform superposition facilitates over-the-air function computation [5], [6] and model aggregation [7], [8].”
Bagci, F., Tegin, B., Kazemi, M., & Duman, T. M. (2026) · doi:10.48550/arxiv.2608.01426
“To address this issue, over-the-air (OTA) computation has been proposed, leveraging the superposition property of the wireless multiple-access channel (MAC) to enable simultaneous aggregation of model updates during transmission [5], [6].”
Machine learning-driven evolution of intelligent unmanned aerial vehicles
Rawat, K. S., & Sharma, T. K. (2026) · Computer Science Review · doi:10.1016/j.cosrev.2026.100937
Phase-Noise-Robust Hierarchical Digital Aircomp for Wireless Federated Learning
Yao, J.-C., Yang, Z.-H., Lin, Y.-J., Shi, W., Xu, J., Huang, Y.-M., & You, X.-H. (2026) · 2026 IEEE/CIC International Conference on Communications in China (ICCC) · doi:10.1109/iccc70295.2026.11680376
“Therefore, integrating AirComp into FL tightly couples uplink transmission with model aggregation, avoiding the substantial spectrum overhead incurred by conventional orthogonal multiple access schemes and enabling low-latency and high spectral efficiency wireless FL [3]–[6].”
Deep Reinforcement Learning in IRS-Assisted Federated Learning
Vesireddy, V., & Naidu, K. (2026) · 2026 International Conference on Emerging Techniques in Computational Intelligence (ICETCI) · doi:10.1109/icetci71772.2026.11710709
“Client indices 0, 11, 16 have almost zero channel gain in direct communication but have extremely high channel gain when RL-controlled IRS link is used.”
Task-aware semantic model-update compression for over-the-air federated learning in 6G edge networks
Al-Zofi, W. (2026) · Telecommunication Systems · doi:10.1007/s11235-026-01500-5
“Despitethisarchitecturaladvantage,therepeatedexchange of high-dimensional model updates remains the central systems bottleneck in practical FL.”
InfluentialAge-of-Information Aware Federated Learning with Finite Speed Pinching Antenna
Wang, K.-D., So, D. K. C., & Ding, Z. (2026) · doi:10.48550/arxiv.2607.23595
“In wireless networks, however, the performance of federated learning is significantly affected by device heterogeneity and communication conditions [2], [3].”
Self-citationLi, Z.-Q., Li, S.-Z., Awada, U., & Zhang, J.-K. (2026) · arXiv.org · doi:10.48550/arxiv.2607.17169
“Jiankang Zhang is with the School of Computing and Engineering, Bournemouth University, UK (Email: jzhang3@bournemouth.ac.uk) Shuangzhi Li is the corresponding author (Email: ielsz@zzu.edu.cn). distortion, which in turn degrades learning performance [11], [12].”
AirMoE: Statistic-Augmented Over-the-Air MoE for Collaborative IntelligencePreprint
Kou, W.-B., Lei, J., Zhu, G., & Yang, Y. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2607.16562
“Subsequent works have refined this paradigm through joint device-selection and beamforming [43].”
FedTrack: Federated Learning for Localization and Velocity Estimation of Moving Targets
Asaad, S., Wang, P., & Tabassum, H. (2026) · IEEE Transactions on Vehicular Technology · doi:10.1109/tvt.2026.3652102
“Nonetheless, noting that the devices are connected to the PS over wireless channels, the aggregation can be directly performed over the air as suggested in [5], [13].”
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.