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 201–300 of 999
GIFAIR-FL: A Framework for Group and Individual Fairness in Federated Learning
Yue, X., Nouiehed, M., & Kontar, R. A. (2022) · INFORMS Journal on Data Science · doi:10.1287/ijds.2022.0022 · Cited by 88
“2 network The need to upload large amounts of data to the cloud incurs high communication and storage costs, demands large internet bandwidth (Jiang et al. 2020, Yang et al. 2020), and leads to latency in deployment as well as reliability risks due to unreliable connection (Zhang et al. 2020c).”
BackgroundJoint Resource Allocation to Minimize Execution Time of Federated Learning in Cell-Free Massive MIMO
Vu, T. T., Ngo, D. T., Ngo, H. Q., Dao, M. N., Tran, N. H. K., & Middleton, R. H. (2022) · IEEE Internet of Things Journal · doi:10.1109/jiot.2022.3183295 · Cited by 63
“It is worth noting that as the UE selection reduces the number of participating UEs, the test accuracy of an FL process is also affected [31].”
BackgroundOver-the-Air Multi-View Pooling for Distributed Sensing
Liu, Z., Lan, Q., Kalør, A. E., Popovski, P., & Huang, K. (2023) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2023.3343086 · Cited by 52
“Most recently, AirComp sees growing popularity in its application to supporting efficient model/gradient aggregation in federated learning, known as over-the-air federated learning [19], [21], [22], [23].”
BackgroundXu, Z., Zhao, D., Liang, W., RANA, O. F., Zhou, P., Li, M., Xu, W., Li, H., & Xia, Q. (2022) · IEEE Transactions on Parallel and Distributed Systems · doi:10.1109/tpds.2022.3218807 · Cited by 62
“Due to the distributive nature of FL, many existing studies promote communication efficiency [34], [41], [57], [63], opti-”
BackgroundDeploying Federated Learning in Large-Scale Cellular Networks: Spatial Convergence Analysis
Lin, Z., Li, X., Lau, V. K. N., Gong, Y., & Huang, K. (2021) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2021.3104834 · Cited by 54
“The versatility and efficiency of over-the-air aggregation has been improved by the development of numerous relevant techniques including digital aggregation [11], gradient compression [10], power control [12], and beamforming [13].”
Wirelessly Powered Federated Edge Learning: Optimal Tradeoffs Between Convergence and Power Transfer
Zeng, Q., Du, Y., & Huang, K. (2021) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2021.3098716 · Cited by 56
“Attempts to overcome the bottleneck have led to the design of a new class of communication techniques for efficient FEEL including over-the-air updates aggregation [5], [8], [9], resource management [10]–[12], adaptive uploading frequency control [4], device scheduling [13], and quantization [14].”
BackgroundZheng, J., Tian, H., Ni, W., Ni, W., & Zhang, P. (2022) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2022.3188502 · Cited by 47
Knowledge-Guided Learning for Transceiver Design in Over-the-Air Federated Learning
Zou, Y., Wang, Z., Chen, X., Zhou, H., & Zhou, Y. (2022) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2022.3192550 · Cited by 48
“On the other hand, the existing studies [11], [14]–[17], [20], [21] mainly adopt optimization-based methods for the transceiver design of AirComp.”
BackgroundMethodInfluentialFederated Learning over Wireless Networks: A Band-limited Coordinated Descent Approach
Zhang, J., Li, N., & Dedeoğlu, M. (2021) · IEEE Conference on Computer Communications (INFOCOM) · doi:10.1109/infocom42981.2021.9488818 · Cited by 53
Federated learning framework for mobile edge computing networks
Fantacci, R., & Picano, B. (2020) · CAAI Transactions on Intelligence Technology · doi:10.1049/trit.2019.0049 · Cited by 52
“The aim of the authors of [2, 21] is the maximisation of the number of devices involved in the aggregation process, by minimising the aggregation error.”
BackgroundMethodInfluentialBabar, M. A., Qureshi, B., & Koubâa, A. (2024) · PLoS ONE · doi:10.1371/journal.pone.0302539 · Cited by 69
The Road to Next-Generation Multiple Access: A 50-Year Tutorial Review
Liu, Y., Ouyang, C., Ding, Z., & Schober, R. (2024) · Proceedings of the IEEE · doi:10.1109/jproc.2024.3476675 · Cited by 53
“Looking ahead, future networks will integrate communication, sensing, and computing (such as mobile edge computing [272] or over-the-air computing [273]) capabilities.”
Self-citationAdvancements in Federated Learning: Models, Methods, and Privacy
Chen, H., Wang, H., Long, Q., Jin, D., & Li, Y. (2025) · ACM Computing Surveys · doi:10.1145/3664650 · Cited by 44
“[154] combine device selection and beamforming design, and investigate the superposition property of a wireless MAC for achieving the fast global model aggregation.”
Lin, Z., Gong, Y., & Huang, K. (2022) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2022.3223661 · Cited by 52
“One particular class of techniques of our interest, known as over-the-air FEEL, features the application of AirComp to realize over-the-air model aggregation in FEEL [11]–[14].”
MethodA Systematic Survey of Blockchained Federated Learning
Wang, Z., Hu, Q., Xu, M., Zhuang, Y., Wang, Y., & Cheng, X. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2110.02182 · Cited by 65
Fine-Grained Data Selection for Improved Energy Efficiency of Federated Edge Learning
Albaseer, A. M., Abdallah, M. M., Al‐Fuqaha, A., & Erbad, A. (2021) · IEEE Transactions on Network Science and Engineering · doi:10.1109/tnse.2021.3100805 · Cited by 63
Over-the-Air Federated Learning and Optimization
Zhu, J., Shi, Y., Zhou, Y., Jiang, C., Chen, W., & Letaief, K. B. (2024) · IEEE Internet of Things Journal · doi:10.1109/jiot.2024.3352280 · Cited by 56
“Prior works [28] demonstrate that AirComp enables fast and spectrum-efficient model aggregation for wireless FL.”
BackgroundMethodInfluentialSelf-citationSynergistic Benefits in IRS- and RS-Enabled C-RAN With Energy-Efficient Clustering
Weinberger, K., Ahmad, A. A., Sezgin, A., & Zappone, A. (2022) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2022.3166393 · Cited by 46
“For this reason, this work proposes the combination of two approaches to find suitable and high quality rank-one solutions: 1) The Gaussian randomization technique [37] and 2) a difference-of-convex (DC) programming approach, which exploits the fact that the nuclear norm and the spectral norm of a positive semidefinit…”
MethodInfluentialDecentralized SGD with Over-the-Air Computation
Özfatura, E., Rini, S., & Gündüz, D. (2020) · Globecom · doi:10.1109/globecom42002.2020.9322286 · Cited by 48
Communication Efficient Federated Learning over Multiple Access Channels
Chang, W.-T., & Tandon, R. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2001.08737 · Cited by 43
Over-the-Air Split Machine Learning in Wireless MIMO Networks
Yang, Y., Zhang, Z., Tian, Y., Yang, Z., Huang, C., Zhong, C., & Wong, K. (2023) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2023.3242701 · Cited by 49
“However, in [18], the scenario is still a multiple access system where all channels are MIMO channels, and the task of OAC is still the weighted sum, which can be regarded as a direct expansion of previous designs in [4], [5], [16], and [17].”
BackgroundInfluentialEnergy Consumption Optimization in RIS-Assisted Cooperative RSMA Cellular Networks
Khisa, S., Elhattab, M., Assi, C., & Sharafeddine, S. (2022) · IEEE Transactions on Communications · doi:10.1109/tcomm.2023.3270907 · Cited by 44
“One possible stopping criterion for problem P7 is given by ||V ||∗−||V ||2 ≤ ζDC, where ζDC > 0 is a sufficiently small constant [30].”
BackgroundMethodInfluentialFederated Spectrum Learning for Reconfigurable Intelligent Surfaces-Aided Wireless Edge Networks
Yang, B. F., Cao, X., Huang, C., Yuen, C., Renzo, M. D., Guan, Y. L., Niyato, D. T., Qian, L., & Debbah, M. A. (2022) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2022.3178445 · Cited by 48
“[30] proposed a fast global model aggregation approach to improve the performance and the convergence rate of FL via over-the-air computation (AirComp).”
BackgroundOver-the-Air Federated Multi-Task Learning Over MIMO Multiple Access Channels
Zhong, C., Yang, H., & Yuan, X. (2022) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2022.3221797 · Cited by 47
“the state-of-the-art OA-FL literature [7], [10], [11].”
BackgroundMethodInfluentialFederated and transfer learning for cancer detection based on image analysis
Bechar, A., Medjoudj, R., Elmir, Y., Himeur, Y., & Amira, A. (2025) · Neural Computing and Applications · doi:10.1007/s00521-024-10956-y · Cited by 51
“The following subsections provide more details on current applications, future directions and limitations of these methods:”
Hou, X., Wang, J., Jiang, C., Meng, Z., Chen, J., & Ren, Y. (2023) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2023.3345393 · Cited by 91
“Moreover, aiming to make full use of the superposition property of multiple-access channels, Yang et al. [14] provided a fast global model aggregation method relying on user selection and beamforming design.”
BackgroundMethodShome, D., Waqar, O., & Khan, W. U. (2022) · Transactions on Emerging Telecommunications Technologies · doi:10.1002/ett.4458 · Cited by 51
“…the scenario of a single cell network where each device transmits a large local A joint beamforming and client selection optimization approach was used by Yang et al. [65] to maximize the number of client devices satisfying the thresholds of satisfactory performance in AirComp based FL training.”
MethodElhattab, M., Arfaoui, M. A., Assi, C. M., & Ghrayeb, A. (2021) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2021.3120989 · Cited by 84
“However, the application of the SDR technique to solve a feasibility check problem during the alternating minimization approach does not assure the feasibility of the obtained solution [42], [43].”
BackgroundMethodInfluentialLiu, Y., Wei, Y., Feng, Z., Zhang, Q., & Han, Z. (2024) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2024.3470873 · Cited by 61
“Motivated by this, it is natural to combine and apply the AirComp and federated learning framework to the vehicular network to realize real-time and efficient ISCC [20].”
InfluentialFedSL: A Communication-Efficient Federated Learning With Split Layer Aggregation
Zhang, W., Zhou, T., Lu, Q., Yuan, Y., Tolba, A. K. R., & Said, W. (2024) · IEEE Internet of Things Journal · doi:10.1109/jiot.2024.3350241 · Cited by 53
“…by transmitting a subset of parameters, instead of all parameters, through low-rank gradient compression [30], [31], [32], [33], random masks [34], [35], etc. Compared to the sparsifica-tion approaches mentioned above, quantization methods like QSGD [36], Atomo [37] and SignSGD [38] propose…”
MethodCommunication-and-Energy Efficient Over-the-Air Federated Learning
Liang, Y., Chen, Q., Zhu, G., Jiang, H., Eldar, Y. C., & Cui, S. R. (2024) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2024.3501297 · Cited by 43
“OTA-FL has emerged as a promising solution for communication-and-energy efficient FL in recent years [24], [25], [26], [27], [28].”
Beamforming Vector Design and Device Selection in Over-the-Air Federated Learning
Kim, M., Swindlehurst, A. L., & Park, D. (2023) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2023.3251339 · Cited by 47
“It increases wireless communication efficiency by aggregating data received simultaneously over the air from distributed devices and thus has been proposed to support efficient model aggregation in FL [15], [16], [17], [18], [19], [20], [21].”
BackgroundMethodInfluentialFederated Learning in Heterogeneous Networks with Unreliable Communication
Zheng, P., Zhu, Y. L., Zhang, Z. J., Hu, Y., & Schmeink, A. (2021) · IEEE Globecom Workshops (GC Wkshps) · doi:10.1109/gcwkshps52748.2021.9682052 · Cited by 20
Broadband Digital Over-the-Air Computation for Wireless Federated Edge Learning
You, L., Zhao, X., Cao, R., Shao, Y., & Fu, L. (2023) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2023.3304652 · Cited by 47
Mu, Y., Garg, N., & Ratnarajah, T. (2022) · IEEE Transactions on Network Science and Engineering · doi:10.1109/tnse.2022.3196463 · Cited by 49
“In this figure, apart from comparing our system’s performance with baseline (green curve) introduced by [10], we also plot the simulation results based on the assumption that DL links are considered to be error-free (black curve) and this assumption has been made in many existing works [37]–[40].”
BackgroundResultDistributed Learning Over a Wireless Network With Non-Coherent Majority Vote Computation
Şahin, A. (2023) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2023.3257712 · Cited by 45
“As compared to the approaches in [25] and [24], the proposed scheme also does not require CSI at the ES or multiple antennas.”
BackgroundResultWorst-Case Design for RIS-Aided Over-the-Air Computation With Imperfect CSI
Zhang, W., Xu, J., Xu, W., You, X., & Fu, W. (2022) · IEEE Communications Letters · doi:10.1109/lcomm.2022.3183869 · Cited by 35
“It has been proved that AirComp can effectively improve communication efficiency and reduce the required bandwidth [4].”
BackgroundTask-Oriented Communication Design in Cyber-Physical Systems: A Survey on Theory and Applications
Mostaani, A., Vu, T. X., Sharma, S. K., Nguyen, V., Liao, Q., & Chatzinotas, S. (2022) · IEEE Access · doi:10.1109/access.2022.3231039 · Cited by 64
“The author in [171] extended the idea to channels with fading channel information at either the transmitter or receiver side.”
BackgroundTime-Sensitive Learning for Heterogeneous Federated Edge Intelligence
Xiao, Y. L., Zhang, X., Li, Y., Shi, G., Krunz, M., Nguyen, D. N., & Hoang, D. T. (2023) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2023.3237374 · Cited by 41
“In some settings, pk can represent the relative portion of the local dataset of each edge server k [4], [35].”
BackgroundDifferentially Private Federated Learning via Reconfigurable Intelligent Surface
Yang, Y., Zhou, Y., Wu, Y., & Shi, Y. (2022) · IEEE Internet of Things Journal · doi:10.1109/jiot.2022.3168066 · Cited by 53
“To solve problem (37), a natural method is to formulate it as a semidefinite programming (SDP) problem by matrix lifting [16], [50].”
BackgroundMethodInfluentialSelf-citationOver-the-Air Federated Learning via Second-Order Optimization
Yang, P., Jiang, Y., Wang, T., Zhou, Y., Shi, Y., & Jones, C. N. (2022) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2022.3185156 · Cited by 41
“that perfect channel state information (CSI) is available at all devices to adjust their transmitted signals based on channel coefficients [26], [30]–[33], [40], [50]–[52].”
BackgroundMethodInfluentialSelf-citationApplying Federated Learning in Software-Defined Networks: A Survey
Ma, X., Liao, L., Li, Z., Lai, R. X., & Zhang, M. (2022) · Symmetry · doi:10.3390/sym14020195 · Cited by 55
“[151] utilized the signal superimposition characteristics of multiple access channels and combined computation and communication to improve communication efficiency.”
MethodAsySQN: Faster Vertical Federated Learning Algorithms with Better Computation Resource Utilization
Zhang, Q., Gu, B., Deng, C., Gu, S., Bo, L., Pei, J., & Huang, H. (2021) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3447548.3467169 · Cited by 30
“Federated learning attracts much attention from both academic and industry McMahan et al. (2017); Yang et al. (2020a); Kairouz et al. (2019); Gong et al. (2016); Zhang et al. (2021) because it meets the emerging demands of collaboratively-modeling with privacy-preserving.”
BackgroundQuyen, N. H., Duy, P. T., Nguyen, N. T., Khoa, N. H., & Pham, V.-H. (2024) · Information Fusion · doi:10.1016/j.inffus.2024.102807 · Cited by 52
Turning Channel Noise Into an Accelerator for Over-the-Air Principal Component Analysis
Zhang, Z., Zhu, G., Wang, R., Lau, V. K. N., & Huang, K. (2022) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2022.3162868 · Cited by 39
“In particular, the mentioned over-the-air FL is a class of techniques that realize over-the-air aggregation by superimposing analog modulated model updates transmitted simultaneously by devices [9], [19]–[23].”
BackgroundHazarika, B., Singh, K. K., Paul, A., & Duong, T. Q. (2023) · IEEE Transactions on Intelligent Vehicles · doi:10.1109/tiv.2023.3335277 · Cited by 49
“Inordertoovercomethe Inordertoovercomethe constraintsimposedbylimitedcommunicationbandwidth,[9] introducedanexpeditedglobalaggregationmethodthrough acombinationofdeviceselectionandbeamformingdesign.”
A Comprehensive Overview of IoT-Based Federated Learning: Focusing on Client Selection Methods
Khajehali, N., Yan, J. H., Chow, Y.-W., & Fahmideh, M. (2023) · Sensors · doi:10.3390/s23167235 · Cited by 48
“After this step, the model is compressed to acquire a sufficient level of performance and to decrease communication costs [63,72,74].”
BackgroundMethodInfluentialConfederated Learning: Federated Learning With Decentralized Edge Servers
Wang, B., Fang, J., Li, H., Yuan, X., & Ling, Q. H. (2023) · IEEE Transactions on Signal Processing · doi:10.1109/tsp.2023.3241768 · Cited by 44
“Due to limited bandwidth, at each iteration only a small subset of users can be selected to interact with the server, which leads to a low efficiency and also calls for a judiciously designed scheduling policy [12], [13], [14].”
BackgroundWang, Y., Guan, P., Yu, H., & Zhao, Y. (2022) · IEEE Access · doi:10.1109/access.2022.3179115 · Cited by 41
“Based on the Proposition 3 in [13], the rank-one constraint (11b) is satisfied when Qt (Qt ) has only one non-zero eigenvalue.”
BackgroundAmplify-and-Forward Relaying for Hierarchical Over-the-Air Computation
Wang, F., Xu, J., Lau, V. K. N., & Cui, S. R. (2022) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2022.3185074 · Cited by 37
“…design objectives (i.e., function computation versus individual data decoding/detection), performance metrics (i.e., the computational MSE [27]–[34] versus the sum MSE of individual users’ data [44], [45]), and solution structures (e.g., the WD-relay-FC channel based phase adaption for…”
BackgroundLi, L., Yang, L., Guo, X., Shi, Y., Wang, H., Chen, W., & Letaief, K. B. (2021) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2021.3118431 · Cited by 37
“Moreover, resource allocation becomes critical to improve the learning convergence rate and model prediction accuracy in wireless FL, including the transmission power control [13]–[16], model aggregation beamforming [17], [18], and device scheduling [18]–[21].”
Self-citationDistributed Intelligence in Wireless Networks
Liu, X., Yu, J., Liu, Y., Gao, Y., Mahmoodi, T., Lambotharan, S., & Tsang, D. H. K. (2023) · IEEE Open Journal of the Communications Society · doi:10.1109/ojcoms.2023.3265425 · Cited by 48
“However, if a large number of users participating in training, it is more challenging to realize reliable aggregation using the wireless multiple access channel in the complex systems [160].”
BackgroundFederated Learning With Over-the-Air Aggregation Over Time-Varying Channels
Tegin, B., & Duman, T. M. (2023) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2023.3235894 · Cited by 40
“In [9], the authors investigate the effects of channel fading on federated learning by employing a worker scheduling algorithm based on channel conditions, while [10] uses multiple receive antennas to maximize the number of selected devices via beamforming.”
BackgroundDistributed Learning over a Wireless Network with FSK-Based Majority Vote
Şahin, A., Everette, B., & Hoque, S. S. M. (2021) · International Conference on Advanced Communication Technologies and Networking (CommNet) · doi:10.1109/commnet52204.2021.9641940 · Cited by 36
“In [12] and [13], blind EDs are considered.”
BackgroundResultFan, X., Wang, Y., Huo, Y., & Tian, Z. (2022) · IEEE Internet of Things Journal · doi:10.1109/jiot.2022.3164339 · Cited by 43
“In multiple antennas scenarios, a joint design of device scheduling and beamforming is presented in [20] to maximize the number of selected workers under the given mean square error (MSE) requirements.”
BackgroundFederated Learning Over Wireless Channels: Dynamic Resource Allocation and Task Scheduling
Chu, S., Li, J., Wang, J., Wang, Z., Ding, M., Zhang, Y., Qian, Y., & Chen, W. (2022) · IEEE Transactions on Cognitive Communications and Networking · doi:10.1109/tccn.2022.3196009 · Cited by 41
“1In general FL, existing work [26] used the accuracy of the test set to measure the performance of machine learning after training.”
MethodUAV-Assisted Covert Federated Learning Over mmWave Massive MIMO
Tong, Z., Wang, J., Hou, X., Jiang, C., & Liu, J. (2024) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2024.3384957 · Cited by 39
“To tackle the limited communication bandwidth issue in the global aggregation, Yang et al. [14] advanced an innovative over-the-air computation (AirComp) based fast aggregation method by harnessing the superposition characteristics of a wireless multiple-access channel.”
MethodToward Scalable Wireless Federated Learning: Challenges and Solutions
Zhou, Y., Shi, Y., Zhou, H., Wang, J., Fu, L., & Yang, Y. (2023) · IEEE Internet of Things Magazine · doi:10.1109/iotm.001.2300099 · Cited by 44
“To facilitate interference management for necessary to develop a cooperative optimization framework to balance the model aggregation accuracy achieved by different FL tasks, thereby enhancing the overall learning performance.”
MethodSelf-citationIRS Assisted Federated Learning: A Broadband Over-the-Air Aggregation Approach
Zhang, D., Xiao, M., Pang, Z., Wang, L., & Poor, H. V. (2024) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2023.3313968 · Cited by 40
“To improve the communication efficiency in model aggregation, over-the-air computation (AirComp) empowered model aggregation approach has emerged [11], [12], [13], [14], [15], [16].”
MethodOver-the-Air Computation Based on Balanced Number Systems for Federated Edge Learning
Şahin, A. (2023) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2023.3320116 · Cited by 35
“Pre-equalization along with digital and analog beamforming are investigated for OAC in [27] and [27], respectively.”
Fast Convergence Algorithm for Analog Federated Learning
Xia, S., Zhu, J., Yang, Y., Zhou, Y., Shi, Y., & Chen, W. (2021) · ICC 2021 - IEEE International Conference on Communications · doi:10.1109/icc42927.2021.9500875 · Cited by 33
“In particular, the authors in [9] proposed a joint device selection and beamforming design to accelerate the convergence of analog federated learning.”
BackgroundInfluentialSelf-citationOnline Optimization for Over-the-Air Federated Learning With Energy Harvesting
An, Q., Zhou, Y., Wang, Z.-B., Shan, H., Shi, Y.-M., & Bennis, M. (2024) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2023.3339298 · Cited by 33
“The transceiver beamforming and device scheduling are jointly optimized for AirComp-assisted FL in [11], while reconfigurable intelligent surface is utilized in [12] and [13] to assist the gradient transmission and hence improve the aggregation accuracy.”
BackgroundMethodSelf-citationUAV-Assisted Multi-Cluster Over-the-Air Computation
Fu, M., Zhou, Y., Shi, Y., Jiang, C., & Zhang, W. (2022) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2022.3227768 · Cited by 35
“A widely adopted performance metric for quantifying the AirComp computation error is MSE between the estimated function value and the ground truth [8], [10], [13].”
InfluentialSelf-citationTo Store or Not? Online Data Selection for Federated Learning with Limited Storage
Gong, C., Zheng, Z., Wu, F., Shao, Y., Li, B., & Chen, G. (2023) · International World Wide Web Conference (WWW) · doi:10.1145/3543507.3583426 · Cited by 38
“the data heterogeneity problem [11, 14, 70, 77], reduce the communication cost [27, 33, 33, 75], select important clients [15, 38, 42, 52] or train a personalized model for each client [20, 72].”
BackgroundBayesian Over-the-Air Computation
Shao, Y., Gündüz, D., & Liew, S. C. (2022) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2022.3229428 · Cited by 36
“That is, symbol-level synchronization among edge devices is assumed and there is only channelgain misalignment in the received signal [6], [16], [21], [23], [29]–[31].”
BackgroundMethodInfluentialOptimizing Over-the-Air Computation in IRS-Aided C-RAN Systems
Yu, D., Park, S.-H., Simeone, O., & Shamai, S. S. (2020) · IEEE International Workshop on Signal Processing Advances in Wireless Communications (SPAWC) · doi:10.1109/spawc48557.2020.9154243 · Cited by 33
“A potential solution to this problem is over-the-air computation (AirComp), which leverages the superposition property of the multiple access channel (MAC) from worker devices to a parameter server (PS) to allow for simultaneous transmissions from multiple devices [2]–[4].”
BackgroundMethodOver-the-Air Federated Learning in Digital Twins Empowered UAV Swarms
Jiang, B., Du, J., Jiang, C., Han, Z., Alhammadi, A., & Debbah, M. A. (2024) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2024.3455009 · Cited by 42
“As an alternative, over-the-air computation (AirComp) found its merits in achieving communication-efficient data aggregation without requiring precise distributed reception in FL systems [20], [21].”
Deep Compression for Efficient and Accelerated Over-the-Air Federated Learning
Khan, F. M. A., Abou-Zeid, H., & Hassan, S. A. (2024) · IEEE Internet of Things Journal · doi:10.1109/jiot.2024.3373460 · Cited by 41
“The PS recovers the noisy global model weights ˆ w , by performing weighted averaging [40] as where R { a } denotes the real part of complex vector a [41] and updates the global model parameters.”
BackgroundImperfect CSI: A Key Factor of Uncertainty to Over-the-Air Federated Learning
Yao, J., Yang, Z., Xu, W., Niyato, D. T., & You, X. (2023) · IEEE Wireless Communications Letters · doi:10.1109/lwc.2023.3318555 · Cited by 34
“In a wireless FL system, the distributed devices, orchestrated by a parameter server (PS), iteratively train a shared learning model through the exchange of model parameters rather than the raw data, thereby protecting data privacy [3], [4].”
BackgroundSpectrum Breathing: Protecting Over-the-Air Federated Learning Against Interference
Wang, Z., Huang, K., & Eldar, Y. C. (2024) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2024.3368197 · Cited by 38
“Building on AirComp, the efficiency of AirFL can be enhanced by beamforming [20], gradient pruning [21], broadband transmission [19] and power control [12], [23], and even the use of Intelligent Reflecting Surface (IRS) [24].”
BackgroundMethodReconfigurable Intelligent Surface for Green Edge Inference in Machine Learning
Hua, S., & Shi, Y. (2019) · IEEE Globecom Workshops (GC Wkshps) · doi:10.1109/gcwkshps45667.2019.9024398 · Cited by 42
“As authors in [16], [24] have pointed out, the performance of SDR degrades sharply once the problem size grows (in our case, M or K becomes large), that’s because the probability of yielding a rank-one solution is low.”
BackgroundSelf-citationEdge Intelligence Oriented Integrated Sensing and Communication: A Multi-Cell Cooperative Approach
Huang, N., Dong, H., Dou, C.-L., Wu, Y., Qian, L.-P., Ma, S.-D., & Lu, R.-X. (2024) · IEEE Transactions on Vehicular Technology · doi:10.1109/tvt.2024.3359094 · Cited by 31
“According to [39], the subgradient of || V km || spec can be computed as v km ( v km ) H , where v km is the eigenvector corresponding to the largest singular value of V km .”
BackgroundDigital Over-the-Air Computation: Achieving High Reliability via Bit-Slicing
Liu, J., Gong, Y., & Huang, K. (2025) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2025.3540455 · Cited by 33
“Researchers have further developed advanced AirFL techniques including power control [14], [15], [16], adaptive precoding [17], scheduling [13], and interference suppression [18].”
Federated Learning With Energy Harvesting Devices
Zeng, L., Wen, D., Zhu, G., You, C., Chen, Q., & Shi, Y. (2023) · IEEE Transactions on Green Communications and Networking · doi:10.1109/tgcn.2023.3310569 · Cited by 32
“Specifically, to improve the communication efficiency, a new analog-communication based multiple access technique, called over-the-air computation, has been proposed [14], [15], [16], [17], [18].”
MethodSelf-citationSumComp: Coding for Digital Over-the-Air Computation via the Ring of Integers
Razavikia, S., Silva, J. M. B. D., & Fischione, C. (2023) · arXiv.org · doi:10.48550/arxiv.2310.20504 · Cited by 31
“The uncoded analog AirComp is extensively studied from various perspectives [4], [5], [12]–[14] and demonstrates significant resource efficiency, particularly in distributed learning, thus garnering attention for federated edge learning systems [15], [16].”
Joint Antenna Selection and Beamforming for Massive MIMO-Enabled Over-the-Air Federated Learning
Asaad, S., Tabassum, H., Ouyang, C., & Wang, P. (2024) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2024.3352504 · Cited by 41
FedFog: Network-Aware Optimization of Federated Learning Over Wireless Fog-Cloud Systems
Nguyen, V., Chatzinotas, S., Ottersten, B., & Duong, T. Q. (2022) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2022.3167263 · Cited by 42
“Various join FL model and radio resource allocation schemes have been proposed in [19]–[21] to minimize either the global loss function or”
BackgroundCommunication Efficient Federated Learning With Energy Awareness Over Wireless Networks
Jin, R., He, X., & Dai, H. (2022) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2021.3138394 · Cited by 66
Age-of-Information Minimization in Federated Learning Based Networks With Non-IID Dataset
Wang, K., Ding, Z., So, D. K. C., & Ding, Z. J. (2024) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2024.3357208 · Cited by 33
“Due to the fact that FL generally involves a large number of devices, device selection is commonly considered to accommodate limited wireless resources or filter unimportant data [17], [18], [19], [20], [21], [22], [23].”
BackgroundSelf-citationExploring Representativity in Device Scheduling for Wireless Federated Learning
Chen, Z., Yi, W., & Nallanathan, A. (2023) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2023.3281765 · Cited by 31
“The device selection and beamforming design in [8] is able to accelerate the convergence speed for over-the-air FL.”
BackgroundJoint Gradient Sparsification and Device Scheduling for Federated Learning
Lin, X., Liu, Y., Chen, F., Ge, X., & Huang, Y. (2023) · IEEE Transactions on Green Communications and Networking · doi:10.1109/tgcn.2023.3239373 · Cited by 39
“The authors in [26] aim to find the maximum selected devices while satisfying the Mean Square Error (MSE) requirement for over the-air federated learning, by jointly optimizing device scheduling and beamforming design.”
BackgroundAceFL: Federated Learning Accelerating in 6G-Enabled Mobile Edge Computing Networks
He, J., Guo, S., Li, M., & Zhu, Y. (2022) · IEEE Transactions on Network Science and Engineering · doi:10.1109/tnse.2022.3190330 · Cited by 37
Over-The-Air Federated Learning Over Scalable Cell-Free Massive MIMO
Sifaou, H., & Li, G. Y. (2023) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2023.3315962 · Cited by 40
“Furthermore, assuming multiple antennas at the CS and designing beam-forming and receiving techniques have been considered in [4], [15], and [19].”
BackgroundInfluentialPerformance-Oriented Design for Intelligent Reflecting Surface-Assisted Federated Learning
Zhao, Y., Wu, Q., Chen, W., Wu, C., & Poor, H. V. (2023) · IEEE Transactions on Communications · doi:10.1109/tcomm.2023.3283799 · Cited by 30
“The existing system design schemes in AirComp-based FL have mainly focused on the separated mean square error (MSE) minimization in each communication round [24], [28], [37].”
MethodOver-the-Air Federated Learning with Joint Adaptive Computation and Power Control
Yang, H.-B., Qiu, P.-Y., Liu, J., & Yener, A. (2022) · International Symposium on Information Theory · doi:10.48550/arxiv.2205.05867 · Cited by 30
Temporal-Structure-Assisted Gradient Aggregation for Over-the-Air Federated Edge Learning
Fan, D., Yuan, X., & Zhang, Y. A. (2021) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2021.3118348 · Cited by 34
“Furthermore, the downlink model broadcasting from the PS to the edge devices is assumed to be error-free by following the convention [10], [17], [18].”
BackgroundMethodGradient Inversion Attacks: Impact Factors Analyses and Privacy Enhancement
Ye, Z., Luo, W., Zhou, Q., Zhu, Z., Shi, Y., & Jia, Y. (2024) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2024.3430533 · Cited by 30
“…is designed to protect privacy by structurally encoding data information, this learning paradigm also faces several challenges, with the most widely studied being issues related to communication [2], [6], security [7], [8], and privacy [9], [10], with privacy being the main focus of this paper.”
BackgroundSelf-citationZhu, M., Zecchin, M., Park, S., Guo, C., Feng, C., & Simeone, O. (2024) · IEEE Transactions on Signal Processing · doi:10.1109/tsp.2024.3358615 · Cited by 36
“Training techniques that operate on noisy channels, as in [4], [41], [42], [43], can be directly accommodated within the proposed federated inference framework.”
BackgroundChen, R., Wan, Q., Zhang, X., Qin, X., Hou, Y., Wang, D., Fu, X., & Pan, M. (2023) · International Conference on Mobile Systems, Applications, and Services · doi:10.1145/3581791.3596865 · Cited by 29
“To reduce FL’s energy consumption, research efforts have been made on network optimization [51], device scheduling [37] and resource utilization optimization [7, 28, 33, 43, 44, 52, 54].”
MethodVertical Federated Learning Over Cloud-RAN: Convergence Analysis and System Optimization
Shi, Y., Xia, S., Zhou, Y., Mao, Y., Jiang, C., & Tao, M. (2023) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2023.3288122 · Cited by 33
“Instead, over-the-air computation (AirComp) serves as a promising approach to enable efficient aggregation for FL [9], [10], [11], [12].”
BackgroundSelf-citationGain Without Pain: Recycling Reflected Energy From Wireless-Powered RIS-Aided Communications
Xie, H., Gu, B.-W., Li, D., Lin, Z., & Xu, Y.-J. (2022) · IEEE Internet of Things Journal · doi:10.1109/jiot.2023.3262517 · Cited by 29
“To tackle this obstacle, we first equivalently transform C8 into the following difference of convex functions constraint [37]”
MethodMulti-Edge Server-Assisted Dynamic Federated Learning With an Optimized Floating Aggregation Point
Ganguly, B., Hosseinalipour, S., Kim, K. T., Brinton, C. G., Aggarwal, V., Love, D. J., & Chiang, M. (2023) · IEEE/ACM Transactions on Networking · doi:10.1109/tnet.2023.3262482 · Cited by 32
“For example, researchers have studied the model training performance under noisy channels [14], limited energy devices [12], limited bandwidth [13], quantization and sparsification [15], [16], and wireless aggregation of signals over the air [17].”
BackgroundRandom Orthogonalization for Federated Learning in Massive MIMO Systems
Wei, X., Shen, C., Yang, J., & Poor, H. V. (2022) · International Conference on Communications · doi:10.1109/icc45855.2022.9838988 · Cited by 13
“MIMO beamforming design to optimize FL has been studied in [6], [7].”
BackgroundMethodCommunication-Efficient ADMM-based Federated LearningPreprint
Zhou, S., & Li, G. Y. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2110.15318 · Cited by 39
“These advantages hint that the proposed algorithms might be practical to deal with many real applications such as mobile edge computing [35, 36], over-the-air computation [37, 38], vehicular communications [1], unmanned aerial vehicle online path control [39] and so forth.”
BackgroundFederated Learning Based Proactive Handover in Millimeter-wave Vehicular Networks
Qi, K., Liu, T., & Yang, C. (2020) · International Conference on Signal Processing · doi:10.1109/icsp48669.2020.9320974 · Cited by 42
“With synchronous FL, the communication cost can be reduced significantly by using over-the-air computation [14], where all users transmit their local models to the MBS simultaneously.”
BackgroundMethodEdge-Assisted Democratized Learning Toward Federated Analytics
Pandey, S. R., Nguyen, M. N. H., Dang, T. N., Tran, N. H. K., Thar, K., Han, Z., & Hong, C. S. (2021) · IEEE Internet of Things Journal · doi:10.1109/jiot.2021.3085429 · Cited by 36
“For example, the authors in [14] implemented an over-the-air computation approach to facilitate fast aggregation via joint device selection and beamforming design.”
MethodDu, M., Zheng, H., Gao, M., Feng, X., Hu, J., & Chen, Y. (2024) · IEEE Internet of Things Journal · doi:10.1109/jiot.2024.3438082 · Cited by 33
“We consider that there are K = 30 edge devices randomly distributed at a distance of [10,200] meters from the BS.”
BackgroundMethodInfluentialOn the Differential Privacy in Federated Learning Based on Over-the-Air Computation
Park, S., & Choi, W. (2023) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2023.3316788 · Cited by 39
“These techniques include communication load reduction [12], [13], [14], analysis of the tradeoff between learning performance and aggregation error [15], power control [16], [17], [18], [19], beamforming [20], and security enhancement [21].”
MethodOver-the-Air Computation with DFT-spread OFDM for Federated Edge Learning
Şahin, A., Everette, B., & Hoque, S. S. M. (2022) · IEEE Wireless Communications and Networking Conference · doi:10.1109/wcnc51071.2022.9771881 · Cited by 29
“As opposed to the methods in [9] and [10], it also does not require CSI at the ES or multiple antennas.”
BackgroundResultOver-the-Air Computation in Correlated Channels
Frey, M., Bjelaković, I., & Stańczak, S. (2021) · IEEE Transactions on Signal Processing · doi:10.1109/tsp.2021.3106115 · Cited by 30
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.