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 601–700 of 999
One-Bit Byzantine-Tolerant Distributed Learning via Over-the-Air Computation
Yang, Y., Wu, Y., Jiang, Y., & Shi, Y. (2023) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2023.3326438 · Cited by 6
“…and computation, over-the-air computation (AirComp) has become a leading analog model aggregation paradigm, which is achieved by exploiting the waveform superposition property of a multiple-access channel, yielding high spectrum efficiency and low transmission latency [30], [31], [32].”
BackgroundMethodSelf-citationTowards Federated Learning on Fresh Datasets
Wu, C., Xiao, M., Wu, J., Xu, Y., Zhou, J., & Sun, H. (2023) · IEEE International Conference on Mobile Ad-Hoc and Sensor Systems (MASS) · doi:10.1109/mass58611.2023.00046 · Cited by 4
“Moreover, we implement the MaxPack algorithm [32] and the Random algorithm for better comparison.”
MethodJoint Power Control and Data Size Selection for Over-the-Air Computation-Aided Federated Learning
An, X., Fan, R., Zuo, S., Hu, H., Jiang, H., & Zhang, N. (2023) · IEEE Internet of Things Journal · doi:10.1109/jiot.2023.3341958 · Cited by 7
“To overcome this problem, over-the-air computation can be used, which is able to directly calculate the summation of uploaded data that are transmitted simultaneously from multiple mobile devices to the BS [10], [11], thanks to the signal superposition feature of the communication channel.”
MethodRevolutionizing Wireless Networks with Federated Learning: A Comprehensive Review
Mahdimahalleh, S. E. (2024) · European Journal of Electrical Engineering and Computer Science · doi:10.24018/ejece.2024.8.6.671 · Cited by 13
“New methods have also been developed that exploit compute-over-air mechanisms and a jointly decode-and-average scheme at the edge computing unit when spectral resources become the communication bottleneck [37], [38].”
Co-Phase Over-the-Air Aggregation for Multi-Server Federated Learning with Randomized Transmissions
Choi, J. (2023) · IEEE Vehicular Technology Conference · doi:10.1109/vtc2023-spring57618.2023.10199823 · Cited by 5
“Thus, unlike [8], no receive beamforming is required.”
BackgroundMethodInfluentialAlaerjan, A. (2023) · Sustainability · doi:10.3390/su15118863 · Cited by 7
“For example, techniques such as over-the-air federated learning provide training capabilities and allow 5G wireless networks to precisely recognize and access edge devices and equipment [11].”
MethodIntegrated Sensing, Communication, and Computation Over-the-Air: Beampattern Design
Wang, S., Li, X., Liu, F., & Gong, Y. (2023) · IEEE International Conference on Communications workshops · doi:10.1109/iccworkshops57953.2023.10283508 · Cited by 7
Louw, S. A., Papilaya, J., & Rehatta, G. (2023) · Journal Markcount Finance · doi:10.55849/jmf.v1i2.86 · Cited by 2
Multicore Federated Learning for Mobile-Edge Computing Platforms
Bai, Y., Chen, L., Li, J., Wu, J., Zhou, P., Xu, Z., & Xu, J. (2022) · IEEE Internet of Things Journal · doi:10.1109/jiot.2022.3224239 · Cited by 6
“[26] maximized the number of selected clients to improve the model performance.”
BackgroundJoint Resource Allocation and Learning Optimization for UAV-Assisted Federated Learning
Liu, C., & Zhu, Q. (2023) · Applied Sciences · doi:10.3390/app13063771 · Cited by 7
He, S., Ren, T., Jiang, X.-H., & Xu, M. (2023) · IEEE Wireless Communications and Networking Conference · doi:10.1109/wcnc55385.2023.10118927 · Cited by 7
“presented an efficient client selection and beamforming algorithm in a communication-constrained environment [7].”
BackgroundJoint Power Control for Over-the-Air Computation in Multicarrier Wireless System
Xie, X. N., Hua, C., & Hong, J. (2022) · Globecom · doi:10.1109/globecom48099.2022.10000853 · Cited by 6
“In [9], [10], AirComp was extended for model aggregation in the distributed edge learning system, which is shown to reduce the communication latency without significant loss of the learning accuracy [11].”
MethodDu, H., Chen, Y., Feng, X., Xiang, Q., & Liu, H. (2022) · Computer Networks · doi:10.1016/j.comnet.2022.109517 · Cited by 8
Barbieri, L., Simeone, O., & Nicoli, M. (2023) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp49357.2023.10095158 · Cited by 9
“We specifically consider the Decentralized Stochastic Gradient Langevin Dynamics (DSGLD) algorithm introduced in [12] under the assumption of noiseless communications, and we propose an analog communication-based implementation that leverages channel-driven sampling and over-the-air computing [15, 16, 17].”
MethodOver-the-Air Computation With Multiple Receivers: A Space-Time Approach
Chen, Z., & Malitsky, Y. (2023) · IEEE Wireless Communications Letters · doi:10.1109/lwc.2023.3275760 · Cited by 5
Best Effort Voting Power Control for Byzantine-resilient Federated Learning Over the Air
Fan, X., Wang, Y., Huo, Y., & Tian, Z. (2022) · IEEE International Conference on Communications workshops · doi:10.1109/iccworkshops53468.2022.9814495 · Cited by 10
“Benefitting from communication-efficient gradient aggregation, FLOA as a cross-disciplinary topic has attracted growing research interests, such as power control [11], [14], [20], devices scheduling [11], [13], [18], gradient compression [10], [12], [16], [17], and beamforming design [15].”
Dynamic Network-Assisted D2D-Aided Coded Distributed Learning
Zeulin, N., Galinina, O., Himayat, N., Andreev, S. D., & Heath, R. W. (2023) · IEEE Transactions on Communications · doi:10.1109/tcomm.2023.3259442 · Cited by 8
“Existing work to reduce the impact of stragglers includes mechanisms such as coded computing [9, 23], over-the-air computing [36, 39], or specific resource allocation [5, 6, 37] and scheduling [29].”
BackgroundFederated Learning: Sum Power Constraints Optimization Design
Chen, Y., Zhang, G., Xu, H., Chen, X., & Li, R. (2021) · Arabian Journal for Science and Engineering · doi:10.1007/s13369-021-06210-5 · Cited by 6
Hierarchical Broadcast Coding: Expediting Distributed Learning at the Wireless Edge
Han, D.-J., Sohn, J.-Y., & Moon, J. J. (2020) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2020.3040792 · Cited by 10
Intelligent reflecting surface-assisted secrecy wireless communication with imperfect CSI
Wang, D., Zhang, J., Zhang, Q., & Wang, H. (2020) · Physical Communication · doi:10.1016/j.phycom.2020.101235 · Cited by 4
Ahmadzadeh, M., Pakravan, S., Hodtani, G. A., Zeng, M., Li, X.-W., Zhang, N., & Chouinard, J.-Y. (2026) · IEEE Transactions on Vehicular Technology · doi:10.1109/tvt.2026.3651441 · Cited by 5
“To meet the demands for low latency and extensive connectivity in FL within IoT networks, a novel approach known as over-the-air FL (OTA-FL) has been introduced [5].”
Xia, Q., Ye, J., Wang, Q., Zou, Y., Lei, C., Chi, Y., Zhou, N., Guerrero, J. M., & García, H. M. (2025) · Applied Energy · doi:10.1016/j.apenergy.2025.127324 · Cited by 5
Energy-Efficient Over-the-Air Federated Learning via Pinching Antenna Systems
Asaad, S., & Bereyhi, A. (2026) · International Conference on Communications · doi:10.1109/icc59461.2026.11587052 · Cited by 3
“Building on this concept, over-the-air federated learning (OTA-FL) has been proposed, where the server uses AirComp to aggregate directly over the air [3], [4].”
Heterogeneous Wireless Federated Learning Framework via Over-the-Air Computation
Xiao, Y., Ye, Y., Tegos, S. A., Diamantoulakis, P. D., You, Y., & Karagiannidis, G. K. (2025) · IEEE Transactions on Vehicular Technology · doi:10.1109/tvt.2025.3600821 · Cited by 5
“This approach is consistent with the federated averaging (FedAvg) framework [12], making the OTA computation-enabled WFL mechanism a promising solution for improving the efficiency of distributed AIoT systems, as explored in [13], [14], [15], [16].”
Over-the-Air Federated Learning: Rethinking Edge AI Through Signal ProcessingPreprint
Azimi‐Abarghouyi, S. M., Fischione, C., & Huang, K. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2512.03719 · Cited by 6
“Nonetheless, approaches based on global CSIT [6] continue to be explored due to their potential for achieving optimal performance.”
InfluentialWang, Y., Zhang, B., Zhang, J., & Li, C. (2025) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2025.3610887 · Cited by 4
“The wireless channels from the edge devices to the PS follow independent and identically distributed (i.i.d.) Rayleigh fading [20], [41].”
OLALa: Online Learned Adaptive Lattice Codes for Heterogeneous Federated LearningPreprint
Lang, N., Simhi, M., & Shlezinger, N. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2506.20297 · Cited by 5
“As a result, a broad range of techniques have been proposed to mitigate communication load, including client selection strategies [7], [8], update sparsification [9]–[11], and over-the-air aggregation schemes [12]–[14].”
Zhao, Y., Xiu, Y., Dai, C., Wei, N., & Niyato, D. (2025) · doi:10.48550/arxiv.2506.00011 · Cited by 6
“Here, G ( n ) denotes the aggregated user gradients (variance at most ˆ α 2 ), and ε ( n ) represents OTA noise with variance at most σ 2 [28].”
Hierarchical Federated Learning With Distributed Clustering and Multichannel ALOHA
Silva, R. V. N. D., dos Reis, A. F., Brante, G. G. D. O., & Souza, R. D. (2025) · IEEE Sensors Journal · doi:10.1109/jsen.2025.3536518 · Cited by 5
“In a different direction, AirComp with FL is attracting a lot of interest in the literature, as in [17], [18], and [19].”
PAUSE: Low-Latency and Privacy-Aware Active User Selection for Federated Learning
Peleg, O., Lang, N., Ben-Ami, D., Rini, S., Shlezinger, N., & Cohen, K. (2025) · IEEE Transactions on Signal Processing · doi:10.1109/tsp.2025.3627252 · Cited by 6
“Communication latency can be tackled by model compression [7], [8], [9], [10], [11], and via over-the-air aggregation in settings where the users share a common wireless channel [12], [13], [14].”
Efficient UAV Swarm-Based Multitask Federated Learning With Dynamic Task Knowledge Sharing
Yang, Y., Yang, T., Wu, X., Guo, Z., & Hu, B. (2026) · IEEE Internet of Things Journal · doi:10.1109/jiot.2026.3672872 · Cited by 9
“For the latter scenario, Li and Liu [14] introduced a UAV object detection task framework based on online FL and adopted over-the-air computation (OTA) [15] technology to improve transmission e ffi ciency.”
Biased Federated Learning Under Wireless Heterogeneity
Abrar, M. F. U., & Michelusi, N. (2026) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2026.3678173 · Cited by 7
“…gradient uploads and the PS decodes all the received local gradients individually followed by their aggregation; and over-the-air FL (OTA-FL) [2], [13]–[18], which exploits the natural superposition property of wireless multiple access channels (MAC) and allow simultaneous transmission to realize…”
InfluentialAge-Based Device Selection and Transmit Power Optimization in Over-the-Air Federated Learning
Liu, J., Chang, Z., & Liang, Y. (2026) · IEEE Transactions on Communications · doi:10.1109/tcomm.2026.3658369 · Cited by 7
“To address this challenge, a low-latency multi-access scheme called over-the-air computation (AirComp) has been proposed [12], [13].”
NCAirFL: CSI-Free Over-the-Air Federated Learning Based on Non-Coherent Detection
Wen, H., Michelusi, N., Simeone, O., & Xing, H. (2025) · International Conference on Communications · doi:10.1109/icc52391.2025.11161902 · Cited by 5
“Other AirFL transmission protocols entailing CSIT include precoding optimization [10] and device selection [11].”
Robust Federated Learning Over the Air: Combating Heavy-Tailed Noise with Median Anchored Clipping
Li, J., Chen, Z., Chong, K. F. E., Das, B., Quek, T. Q. S., & Yang, H. H. (2025) · International Symposium on Modeling and Optimization in Mobile, Ad Hoc, and Wireless Networks (WiOpt) · doi:10.23919/wiopt66569.2025.11123216 · Cited by 7
Uplink Over-the-Air Aggregation for Multi-Model Wireless Federated Learning
Zhang, C., Dong, M., Liang, B., Afana, A., & Ahmed, Y. (2024) · International Workshop on Signal Processing Advances in Wireless Communications · doi:10.1109/spawc60668.2024.10694148 · Cited by 5
“Various design schemes have been proposed to improve the communication efficiency of wireless FL, including transmission design of the downlink [2], uplink [3]–[5], and combined downlink-uplink [6]–[8].”
BackgroundOver-the-Air Federated Learning Client Selection in Integrated Sensing, Computing and Communication
Zheng, P., Zhu, Y. P., Yulin, H., & Schmeink, A. (2024) · IEEE International Conference on Communications workshops · doi:10.1109/iccworkshops59551.2024.10615482 · Cited by 3
“Balanced dataset sizes are assumed throughout the rest of this work as in [2], [8], i.e.,”
BackgroundInfluentialMixed-Precision Federated Learning via Multi-Precision Over-the-Air Aggregation
Yuan, J., Wei, Z., & Guo, W. (2025) · IEEE Wireless Communications and Networking Conference · doi:10.1109/wcnc61545.2025.10978511 · Cited by 4
“Recent advancements in FL research [2] primarily focus on two key aspects: (i) enhancing privacy and security, and (ii) improving efficiency.”
An Autoencoder-Based Constellation Design for AirComp in Wireless Federated Learning
Mu, Y., Wei, X., & Shen, C. (2024) · International Conference on Communications · doi:10.1109/icc51166.2024.10622489 · Cited by 6
“As a result, it is commonly assumed that the download communication phase is error-free [4], [5], [11]– [13].”
BackgroundBiased Over-the-Air Federated Learning under Wireless Heterogeneity
Abrar, M. F. U., & Michelusi, N. (2024) · IEEE International Conference on Communications workshops · doi:10.1109/iccworkshops59551.2024.10615488 · Cited by 5
“Recently, OTA computation has emerged as a promising candidate over conventional digital communication approaches for realizing FL solutions that are device-scalable [7], [8], [14].”
BackgroundMethodInfluentialWang, J.-B., Wang, B., Ding, C., Lin, M., & Wang, J. (2023) · IEEE Transactions on Green Communications and Networking · doi:10.1109/tgcn.2023.3325385 · Cited by 4
“Compared with the optimization of some of its variables, the federated learning algorithm assisted by wireless computing was envisaged while the user privacy has been well protected, and the delay and power consumption were reduced [22].”
BackgroundMulti-Model Wireless Federated Learning with Downlink Beamforming
Zhang, C., Dong, M., Liang, B., Afana, A., & Ahmed, Y. (2023) · IEEE International Conference on Acoustics, Speech, and Signal Processing · doi:10.1109/icassp48485.2024.10445815 · Cited by 5
“Assuming an error-free downlink, [3]–[8] focused on efficient transmission for the uplink acquisition of local parameters from devices to the BS, including both digital [3] and analog [4]–[8] schemes.”
MethodFederated Learning via Active RIS Assisted Over-the-Air Computation
Zhang, D., Xiao, M., Skoglund, M., & Poor, H. V. (2024) · IEEE International Conference on Machine Learning for Communication and Networking (ICMLCN) · doi:10.1109/icmlcn59089.2024.10624924 · Cited by 6
“Existing works in this area avoided large aggregation errors mainly by excluding “stragglers”, i.e., devices with weak channels, from concurrent model uploading [6], [7].”
BackgroundMovable Antennas-Assisted Over-the-Air Computation: Dynamic and Static Design
Cheng, Z.-Q., Li, N.-X., Zhu, J.-C., Yang, S., Ouyang, C.-J., & Zhang, X.-Q. (2023) · doi:10.48550/arxiv.2309.12596 · Cited by 5
“…node (SN) collects a measurement of a parameter of interest, such as temperature or humidity, and simultaneously transmits it to a central fusion center over a shared wireless channel, where the desired function (e.g., arithmetic mean or sum) is computed directly from the superimposed signals [6].”
Over-the-Air Computation Assisted Hierarchical Personalized Federated Learning
Zhou, F., Wang, Z., Luo, X., & Zhou, Y. (2023) · International Conference on Communications · doi:10.1109/icc45041.2023.10278799 · Cited by 8
“Such a feature can be leveraged to achieve low-latency model aggregation in FL, as the server only requires the average of local models to obtain an updated global model [9]–[14].”
BackgroundDong, Y., Wang, L., Chi, Y., Wang, J., Zhang, H., Yu, F., Leung, V. C. M., & Hu, X. (2023) · arXiv.org · doi:10.48550/arxiv.2303.17885 · Cited by 5
“The signal encoding algorithms were designed for the analog WFL system [30], [35], [36].”
MethodFederated Edge Learning via Reconfigurable Intelligent Surface with One-Bit Quantization
Li, H., Wang, R., Wu, J. H., & Zhang, W. (2022) · Globecom · doi:10.1109/globecom48099.2022.10001550 · Cited by 6
“The similar solution process can be found in [11], which is skipped due to space limitation.”
BackgroundMethodOver-the-Air Computing under Adaptive Channel State Estimation
Evgenidis, N. G., Papanikolaou, V., Diamantoulakis, P. D., & Karagiannidis, G. (2022) · 2022 18th International Conference on Wireless and Mobile Computing, Networking and Communications (WiMob) · doi:10.1109/wimob55322.2022.9941709 · Cited by 5
Quick and Reliable Physical-layer Data Aggregation in LoRa through Multi-Packet Reception
Tang, Z., You, L., Dai, H., & Fu, L. (2022) · IEEE International Conference on Sensing, Communication, and Networking (SECON) · doi:10.1109/secon55815.2022.9918167 · Cited by 6
“[20] and [21] consider OFDM-based PDA for federated edge learning.”
BackgroundA Review on Federated Learning approach in Artificial Intelligence
Govindwar, G. D., & Dhande, S. S. (2022) · International Conference on Computing Communication Control and automation · doi:10.1109/iccubea54992.2022.10010798 · Cited by 4
“[10] proposed, centralized vertical federated logical regression approach that achieves logical regression technique in vertical learning, using the following function.”
BackgroundCommunication-Efficient Federated Learning For Massive MIMO Systems
Mu, Y., Garg, N., & Ratnarajah, T. (2022) · IEEE Wireless Communications and Networking Conference · doi:10.1109/wcnc51071.2022.9771851 · Cited by 5
“While many of these studies only consider the imperfections introduced by UL transmissions [15]–[17], effects of DL transmissions are rarely investigated; that is, the fact that the aggregated model being broadcast in DL links also experiences fading and interference, has not been taken into…”
Timed Loops for Distributed Storage in Wireless Networks
Mukherjee, A., Deb, P. K., & Misra, S. (2022) · IEEE Transactions on Parallel and Distributed Systems · doi:10.1109/tpds.2021.3100780 · Cited by 5
“Although the local ground-level IoT networks, within the purview of a gateway, may be well connected, the last-mile connectivity from the gateway to the cellular service provider’s infrastructure is often constrained [3] or absent (as shown in Fig.”
BackgroundFLOM: Toward Efficient Task Processing in Big Data with Federated Learning
Wu, C.-Y., & Li, Y. (2022) · Security and Communication Networks · doi:10.1155/2022/5277362 · Cited by 5
RetractedAnalog MIMO Communication for One-Shot Distributed Principal Component Analysis
Chen, X., Larsson, E. G., & Huang, K. (2022) · IEEE Transactions on Signal Processing · doi:10.1109/tsp.2022.3182484 · Cited by 4
“Two solutions warranting further investigation are to use AirAggregate [18], [25], [26] techniques and to equip the server with a large-scale array that can be used to resolve multiple spatially multiplexed transmissions from the devices [39].”
MethodRemoving Channel Estimation by Location-Only Based Deep Learning for RIS Aided Mobile Edge Computing
Hu, X., Masouros, C., & Wong, K. (2021) · International Conference on Communications · doi:10.1109/icc42927.2021.9500961 · Cited by 6
“Note that one subgradient of ||Ψ||s at point Ψ can be efficiently computed as x1x1 , where x1 is the vector corresponding to the largest singular value of Ψ [24].”
BackgroundMAC Aware Quantization for Distributed Gradient Descent
Chang, W.-T., & Tandon, R. (2020) · Globecom · doi:10.1109/globecom42002.2020.9322254 · Cited by 7
Scheduling Policies for Federated Learning in Wireless Networks: An Overview
Shi, W., Sun, Y., Huang, X.-F., Zhou, S., & Niu, Z.-S. (2020) · doi:10.12142/ztecom.202002003 · Cited by 5
Over-the-Air Computation via Cloud Radio Access Networks
Xing, L., Zhou, Y., & Shi, Y. (2021) · 2021 IEEE International Conference on Communications Workshops (ICC Workshops) · doi:10.1109/iccworkshops50388.2021.9473527 · Cited by 4
Self-citationPakravan, S., Trigui, I., Ajib, W., & Zhu, W. (2026) · doi:10.48550/arxiv.2604.12154 · Cited by 4
“To address this issue, we exploit its difference-of-convex (DC) structure and solve it within the DC programming framework [15].”
Qian, B., Qian, J., Wen, Z., Wu, D., He, S., Chen, J., & Ranjan, R. K. (2025) · IEEE Transactions on Cognitive Communications and Networking · doi:10.1109/tccn.2025.3547049 · Cited by 4
“Over-the-air (OTA) updates play a crucial role in distributed deep learning applications, ensuring seamless integration of the latest advancements and improvements, and empowering the models to continuously evolve and adapt to emerging system failures [23], [24].”
Non-Convex Over-the-Air Heterogeneous Federated Learning: A Bias–Variance Trade-off
Abrar, M. F. U., & Michelusi, N. (2026) · International Conference on Communications · doi:10.1109/icc59461.2026.11587568 · Cited by 4
“Among these approaches, over-the-air (OTA) computation has emerged as a particularly scalable solution for wireless FL. OTA-FL works on the principle of waveform superposition on the multiple-access channel (MAC), whereby simultaneous transmissions yield “single-shot” parameter aggregation at the…”
InfluentialZheng, S., Zhang, L., Dong, Y., Li, X., Wang, X., & Chen, X. (2025) · IEEE Internet of Things Journal · doi:10.1109/jiot.2025.3591626 · Cited by 5
“Increasing M generally improves learning by utilizing more data [27], [35].”
Communication-Efficient Personalized Distributed Learning With Data and Node Heterogeneity
Tian, Z., Zhang, Z., Li, Y., & Bennis, M. (2025) · IEEE Transactions on Cognitive Communications and Networking · doi:10.1109/tccn.2025.3564354 · Cited by 5
“Consequently, there have been extensive works aiming to reduce the communication cost through resource allocation, over-the-air computation [26], accelerating convergence [27], and minimizing communication cost per iteration [28], [29], [30], [31].”
PAUSE: Privacy-Aware Active User Selection for Federated Learning
Peleg, O., Lang, N., Rini, S., Shlezinger, N., & Cohen, K. (2025) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp49660.2025.10890399 · Cited by 4
“Communication latency can be tackled by model compression [6]–[9], and via over-the-air aggregation when the users share a wireless channel [10]–[12].”
Update Estimation and Scheduling for Over-the-Air Federated Learning with Energy Harvesting Devices
Bagci, F., Tegin, B., Kazemi, M., & Duman, T. M. (2025) · 2025 IEEE International Conference on Communications Workshops (ICC Workshops) · doi:10.1109/iccworkshops67674.2025.11162338 · Cited by 4
“However, FL relies on iterative transmission of local updates from the mobile users (MUs) to the parameter server (PS), making communication band-width a key bottleneck [3].”
Fed-ZOE: Communication-Efficient Over-the-Air Federated Learning via Zeroth-Order Estimation
Jang, J., Lyu, H., Love, D. J., & Yang, H. J. (2026) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2026.3739217 · Cited by 4
“In the context of FedAvg [8], the function f becomes a weighted sum of the local updates ∆ w i , aligning perfectly with the model aggregation process in FL where local updates are summed over devices [6].”
InfluentialClipper : Online Joint Client Sampling and Power Allocation for Wireless Federated Learning
Xu, W., Liang, B., Boudreau, G., & Sökün, H. Ü. (2024) · ACM Transactions on Modeling and Performance Evaluation of Computing Systems · doi:10.1145/3703628 · Cited by 6
“Finally, we note that time-varying channels were also widely considered in FL with over-the-air computation [34].”
Li, Z.-J., Wu, H., Lu, Y.-L., Dai, Y.-Y., & Ai, B. (2024) · China Communications · doi:10.23919/jcc.fa.2023-0180.202408 · Cited by 4
“…based on the traditional aggregation scheme, such as parameter compression techniques (e.g., quantization and subsampling) [16, 17], client selection [18], broad-band analog aggregation with over-the-air computation [19], and digital twin techinique [20], to reduce the occurrences of stragglers.”
MethodOver-the-Air Computation Assisted Federated Learning with Progressive Training
Gao, G., An, Q., Wang, Z., Wang, Z., Shi, Y., & Zhou, Y. H. (2024) · International Conference on Communications · doi:10.1109/icc51166.2024.10622804 · Cited by 4
“Therefore, various approaches were proposed to mitigate the transmission distortion via power control [11], [12], beamforming [13], [14], scheduling [15].”
BackgroundSelf-citationSecrecy Rate Optimization for STAR-RIS-Enhanced UAV Communications
Zhao, Y., Zhang, Q., Hai, L., Song, Z., & Hou, S. (2024) · International Conference on Communications · doi:10.1109/icc51166.2024.10622268 · Cited by 5
“Then, the objective function (29a) can be recast as η − υ ∥ 2 can be efficiently computed as ee H , where e is the eigenvector corresponding to the largest eigenvalue of E pn ( τ − 1) [13].”
MethodAdaptive Dataset Management Scheme for Lightweight Federated Learning in Mobile Edge Computing
Kim, J., Bang, J., & Lee, J. (2024) · Sensors · doi:10.3390/s24082579 · Cited by 4
Stochastic gradient compression for federated learning over wireless network
Lin, X., Liu, Y., Chen, F.-J., Huang, Y., & Ge, X.-H. (2024) · China Communications · doi:10.23919/jcc.fa.2022-0660.202404 · Cited by 4
“The authors in [26] present the federated learning scheme with over-the-air computation, and jointly optimize device selection and beamforming design.”
BackgroundBoosting Fairness and Robustness in Over-the-Air Federated Learning
Öksüz, H. Y., Molinari, F., Sprekeler, H., & Raisch, J. (2024) · IEEE Control Systems Letters · doi:10.1109/lcsys.2024.3402123 · Cited by 4
“Unlike the existing literature on Over-the-Air computation, e.g., [8], [17], the proposed algorithm can operate despite the inherently unknown nature of channel coefficients.”
Random Aggregate Beamforming for Over-the-Air Federated Learning in Large-Scale Networks
Xu, C., Zhang, C., Huang, Y., & Niyato, D. T. (2024) · IEEE Internet of Things Journal · doi:10.1109/jiot.2024.3360190 · Cited by 9
“1) Novel DC Method [31]: The DC representation is adopted to induce the sparsity.”
BackgroundMethodInfluentialMhanna, E., & Assaad, M. (2024) · Allerton Conference on Communication, Control, and Computing · doi:10.1109/allerton63246.2024.10735291 · Cited by 8
“In [15] and [26] and all references within, they assume that they can remove the impact of the channel.”
BackgroundEnergy-Efficient and Fast Controlled Descent for Over-the-Air Assisted Federated Learning
Adhikary, S., & Mehta, N. B. (2023) · Globecom · doi:10.1109/globecom54140.2023.10437140 · Cited by 4
AirFL-Mem: Improving Communication-Learning Trade-Off by Long-Term Memory
Wen, H., Xing, H., & Simeone, O. (2024) · IEEE Wireless Communications and Networking Conference · doi:10.1109/wcnc57260.2024.10571203 · Cited by 4
“To provide additional context for this work, other solutions to the problem of deep fading in AirFL in the presence of CSIT include phase-only compensation [8], [9], transmission weight optimization [10], and device selection [11].”
MethodSWAP: Sparse Entropic Wasserstein Regression for Robust Network Pruning
You, L., & Cheng, H. V. (2023) · International Conference on Learning Representations · doi:10.48550/arxiv.2310.04918 · Cited by 4
“Additionally, over-the-air communications itself suffer from unavoidable noises (Ang et al., 2020; Yang et al., 2020).”
BackgroundCommunication-Efficient Vertically Split Inference via Over-the-Air Computation
Yang, P., Wen, D., Zeng, Q., Wang, T., & Shi, Y. (2023) · IEEE International Workshop on Signal Processing Advances in Wireless Communications (SPAWC) · doi:10.1109/spawc53906.2023.10304428 · Cited by 7
“To resolve the communication bottleneck, we propose to adopt the technique of over-the-air computation (AirComp), which has shown its effectiveness for low-latency aggregation in many research [17], [18].”
MethodSelf-citationJoint Trajectory and Communication Optimization for UAV-Assisted Over-The-Air Federated Learning
Hsu, K., Lee, M., & Hong, Y.-W. P. (2023) · IEEE International Conference on Communications workshops · doi:10.1109/iccworkshops57953.2023.10283588 · Cited by 7
“On the other hand, to further enhance the FL framework in terms of resource utilization, over-the-air (OTA) aggregation [10] for computation was introduced in [7]–[9] by exploiting the superposition property of the wireless channel.”
BackgroundConvergence Analysis of Over-the-Air FL with Compression and Power Control via Clipping
Wen, H., Xing, H., & Simeone, O. (2023) · Globecom · doi:10.1109/globecom54140.2023.10437291 · Cited by 4
“The resulting protocols, referred to as AirFL here, have been studied, for instance, in [2], which proposed an optimization scheme for device selection and beamforming to minimize the aggregation error.”
BackgroundMethodInfluentialFederated Learning in Wireless Networks via Over-the-Air Computations
Öksüz, H. Y., Molinari, F., Sprekeler, H., & Raisch, J. (2023) · IEEE Conference on Decision and Control (CDC) · doi:10.1109/cdc49753.2023.10384001 · Cited by 5
“Due to its advantages, federated learning has been carried out by [23]–[25] via over-the air computation.”
BackgroundPrivacy-Preserving Object Detection With Poisoning Recognition for Autonomous Vehicles
Li, J., Guo, W., Xie, L., Liu, X., & Cai, J. (2022) · IEEE Transactions on Network Science and Engineering · doi:10.1109/tnse.2022.3227119 · Cited by 4
Transform-Domain Federated Learning for Edge-Enabled IoT Intelligence
Zhao, L., Cai, L., & Lu, W.-S. (2022) · IEEE Internet of Things Journal · doi:10.1109/jiot.2022.3222842 · Cited by 5
Toward Secure and Private Over-the-Air Federated LearningPreprint
Yan, N., Wang, K., Zhi, K., Pan, C., Chai, K. K., & Poor, H. V. (2022) · arXiv (Cornell University) · doi:10.48550/arxiv.2210.07669 · Cited by 4
Content-Aware Client Selection for Federated Learning in Wireless Networks
Qiao, Z., Shen, Y., Yu, X.-H., Zhang, J., Song, S.-H., & Letaief, K. (2022) · International Mediterranean Conference on Communications and Networking · doi:10.1109/meditcom55741.2022.9928665 · Cited by 4
Utility Optimization for Over-the-air Computation Systems with Spectrum Sharing
Li, F., Ye, Q., Fapi, E. T., Sun, W., & Jiang, Y. (2022) · International Conference on Communications · doi:10.1109/icc45855.2022.9838909 · Cited by 4
“The maximization of the number of scheduled users is addressed for a federated learning system with SIMO setting in [5].”
BackgroundFederated Learning: An Energy Efficiency Perspective to Extend Smartphone Battery Life
Herle, A., & Channegowda, J. (2022) · IEEE Delhi Section Conference (DELCON) · doi:10.1109/delcon54057.2022.9753113 · Cited by 4
“The server is then tasked to combine these separate models together into a ’global model’ [8].”
MethodFederated Learning over Time-Varying Channels
Tegin, B., & Duman, T. M. (2021) · Globecom · doi:10.1109/globecom46510.2021.9685486 · Cited by 5
“In [8], the authors investigate the effects of channel fading on federated learning where they employ a worker scheduling algorithm based on the specific channel conditions, while [9] uses multiple receive antennas to maximize the number of selected devices via beamforming.”
BackgroundCorrelated Participation Decision Making for Federated Edge Learning
Hu, Q., Li, F., Zou, X., & Xiao, Y. (2020) · Globecom · doi:10.1109/globecom42002.2020.9321981 · Cited by 6
User Selection Aware Joint Radio-and-Computing Resource Allocation for Federated Edge Learning
Zuo, Y., & Liu, Y. (2020) · International Conference on Wireless Communications and Signal Processing (WCSP) · doi:10.1109/wcsp49889.2020.9299802 · Cited by 5
“User selection criteria in [11] are designed by low-rank optimization and beamforming based on channel-state information (CSI).”
BackgroundChen, X., Cai, Y., Liu, A., & Hanzo, L. (2021) · IEEE Transactions on Vehicular Technology · doi:10.1109/tvt.2021.3112930 · Cited by 4
“To this end, we devise a novel technique of enhancing the sparsity structures in the solutions for problem P1 with the aid of the Ky Fan n-norm of [13], [21].”
BackgroundMethodEnabling Intelligence at Network Edge:An Overview of Federated Learning
Howard, H., Zhao, Z.-Y., & Quek, T. Q. S. (2020) · doi:10.12142/ztecom.202002002 · Cited by 4
Joint Annotator-and-Spectrum Allocation in Wireless Networks for Crowd Labeling
Li, X., Zhu, G., Shen, K., Yu, W., Gong, Y., & Huang, K. (2020) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2020.3000241 · Cited by 7
“One main focus in this area is on communication efficient techniques for acquiring and leveraging distributed data for AI model training [3]–[5].”
MethodDistributed Computing System on a Smartphones-Based Network
Salem, H. (2019) · Lecture notes in computer science · doi:10.1007/978-3-030-29852-4_26 · Cited by 9
Sense Smarter, Think Better: A Survey on Edge Perception for Next-Generation NetworksPreprint
Lyu, Z., Cao, X., Song, X., Li, Y., Wang, J., Wang, S., Cui, Y., Yuan, W., Yu, X., Zhu, G., et al. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2605.18457 · Cited by 7
“For example, [101] jointly optimizes user selection and receive beamforming to maximize device participation under a computation MSE constraint, while [102] proposes truncated power control to exclude deeply faded devices, balancing learning performance and aggregation error.”
Coherence-Aware Over-the-Air Distributed Learning Under Heterogeneous Link Impairments
Karbalayghareh, M., Love, D. J., & Brinton, C. G. (2026) · IEEE Journal on Selected Areas in Information Theory · doi:10.1109/jsait.2026.3672794 · Cited by 4
“The second is over-the-air federated learning (OTA-FL), which exploits the natural superposition property of the wireless multiple-access channel to perform simultaneous analog transmissions, enabling one-shot gradient aggregation [24], [25], [26], [27].”
Huang, N., Li, P.-C., Qian, L.-P., Ren, Y.-Z., & Wu, Y. (2026) · IEEE Internet of Things Journal · doi:10.1109/jiot.2025.3625450 · Cited by 3
“…(61) and (62) into two linear constraints as follows: In (50), the rank-one property of V rad kn leads to the following constraint: According to [32], there exists the equivalence between the rank-one constraint (65) and the following DC form: where || • || spec denotes the spectral norm.”
Adaptive Federated Learning for energy-aware wireless edge networks in 6G environments
Mohammed, B. (2026) · Computer Communications · doi:10.1016/j.comcom.2026.108452 · Cited by 3
Towards Optimal Constellation Design for Digital Over-the-Air ComputationPreprint
Razavikia, S., Gündüz, D., & Fischione, C. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2511.06372 · Cited by 2
“OAC thereby narrows the gap between communication and computation and has enabled diverse applications, including federated learning [7]–[10], distributed inference [11], on-chip computing [12], and wireless control [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.