Joint Device Scheduling and Resource Allocation for Latency Constrained Wireless Federated Learning
Open accessThe DOI matches its Crossref record; 36 of the 36 references it cites were found in a registry or an index.
Abstract
In federated learning (FL), devices contribute to the global training by uploading their local model updates via wireless channels. Due to limited computation and communication resources, device scheduling is crucial to the convergence rate of FL. In this paper, we propose a joint device scheduling and resource allocation policy to maximize the model accuracy within a given total training time budget for latency constrained wireless FL. A lower bound on the reciprocal of the training performance loss, in terms of the number of training rounds and the number of scheduled devices per round, is derived. Based on the bound, the accuracy maximization problem is solved by decoupling it into two sub-problems. First, given the scheduled devices, the optimal bandwidth allocation suggests allocating more bandwidth to the devices with worse channel conditions or weaker computation capabilities. Then, a greedy device scheduling algorithm is introduced, which selects the device consuming the least updating time obtained by the optimal bandwidth allocation in each step, until the lower bound begins to increase, meaning that scheduling more devices will degrade the model accuracy. Experiments show that the proposed policy outperforms state-of-the-art scheduling policies under extensive settings of data distributions and cell radius.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
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
“The trade-off between the convergence speed and the per-round latency was revealed in [242] based on the key observation that more scheduled devices yield faster convergence rate while prolonging the time of uploading the local updates at each iteration due to limited radio resources.”
From Edge artificial intelligence for 6g: vision, enabling technologies, and applications (2022) · cites this paper as [242]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Gradient-based learning applied to document recognition
Lecun, Y., Bottou, L., Bengio, Y., & Haffner, P. (1998) · Proceedings of the IEEE · doi:10.1109/5.726791 · Cited by 58,805 · cited 1×
“We evaluate the training performance of the proposed policy under two well-known learning tasks, the MNIST dataset [31] for handwritten digits classification and the CIFAR10 dataset [32] for image classification.”
MethodCommunication-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 1×
“To address the aforementioned issues, researchers have proposed a new distributed model training framework called Federated Learning (FL) [6], [7].”
BackgroundFederated learning: Challenges, methods, and future directions
Li, T., Sahu, A. K., Talwalkar, A., & Smith, V. (2020) · IEEE Signal Processing Magazine · doi:10.1109/msp.2020.2975749 · Cited by 6,210
A Vision of 6G Wireless Systems: Applications, Trends, Technologies, and Open Research Problems
Saad, W., Bennis, M., & Chen, M. (2020) · IEEE Network · doi:10.1109/mnet.001.1900287 · Cited by 4,985 · cited 1×
“Since uploading raw data via wireless channels can drain the wireless bandwidth and cause privacy issues when the raw data are uploaded to the central server [4], it is hardly practical to use conventional centralized training methods in wireless networks [5].”
MethodFederated Learning in Mobile Edge Networks: A Comprehensive Survey
Lim, W. Y. B., Luong, N. C., Hoang, D. T., Jiao, Y., Liang, Y.-C., Yang, Q., Niyato, D., & Miao, C. (2020) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2020.2986024 · Cited by 2,565 · cited 1×
“A typical wireless FL system leverages the computation capabilities of multiple end devices, which are coordinated by a central controller, for example a base station (BS), to train a model in an iterative fashion [8].”
BackgroundEdge Intelligence: Paving the Last Mile of Artificial Intelligence With Edge Computing
Zhou, Z., Chen, X., Li, E., Zeng, L., Luo, K., & Zhang, J. (2019) · Proceedings of the IEEE · doi:10.1109/jproc.2019.2918951 · Cited by 2,455 · cited 1×
“Therefore, the model accuracy within certain training time budgets (i.e., convergence rate w.r.t. time ), which is critical for latency constrained FL applications [27], [28], has not been addressed yet.”
BackgroundTowards 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 1×
“Therefore, FL becomes a promising technology for distributed data analysis and model training in wireless networks [9], [10], and has been used in many applications, for instance, resource allocation optimization in vehicle-to-vehicle (V2V) communications [11] and content recommendations for smartphones [12].”
BackgroundAdaptive Federated Learning in Resource Constrained Edge Computing Systems
Wang, S., Tuor, T., Salonidis, T., Leung, K. K., Makaya, C., He, T., & Chan, K. (2019) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2019.2904348 · Cited by 2,232 · cited 10×
“Based on Lemma 3 in [20], we have ‖wi,k − vk‖ ≤ gi(τ) and ‖wj,k−vk‖ ≤ gj(τ), where gi(x) δi β ((ηβ + 1) − 1).”
BackgroundMethodInfluentialClient Selection for Federated Learning with Heterogeneous Resources in Mobile Edge
Nishio, T., & Yonetani, R. (2019) · ICC 2019 - 2019 IEEE International Conference on Communications (ICC) · doi:10.1109/icc.2019.8761315 · Cited by 1,875 · cited 3×
“the number of rounds, which is consistent with conclusions from existing work [23], [24].”
BackgroundResultInfluentialClient Selection for Federated Learning with Heterogeneous Resources in Mobile Edge
Nishio, T., & Yonetani, R. (2018) · arXiv · doi:10.48550/arxiv.1804.08333 · Cited by 1,875 · cited 3×
“the number of rounds, which is consistent with conclusions from existing work [23], [24].”
BackgroundResultInfluential
Research works printed but not linked: 4
- [32] A. Krizhevsky, V. Nair, and G. Hinton, 'The CIFAR-10 dataset,' online: http://www. cs. toronto. edu/kriz/cifar. html , 2014. -
- [11] S. Samarakoon, M. Bennis, W. Saad, and M. Debbah, 'Federated learning for ultra-reliable low-latency v2v communications,' in 2018 IEEE Global Communications Conference (GLOBECOM) , Abu Dhabi, United Arab Emirates, 2018, pp. 1-7. - · DOI · Paper page
- [3] I. Goodfellow, Y. Bengio, A. Courville, and Y. Bengio, Deep learning . MIT press Cambridge, 2016, vol. 1. -
- [2] Cisco global cloud index: Forecast and methodology. [Online]. Available: https://www.cisco.com/c/en/us/solutions/ collateral/service-provider/global-cloud-index-gci/white-paper-c11-738085.html -
Cited by
Items 101–200 of 380
Asynchronous Wireless Federated Learning With Probabilistic Client Selection
Yang, J., Liu, Y., Chen, F., Chen, W., & Li, C. (2023) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2023.3337773 · Cited by 31
“For example, the authors in [29] consider joint client selection and bandwidth allocation to maximize the model accuracy obtained in a limited training time.”
BackgroundInfluentialLi, Y., Wu, Y., Song, Y., Qian, L., & Jia, W. (2022) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2022.3201622 · Cited by 34
“In [20], a joint policy of device-scheduling and resource allocation has been proposed to maximize the FL accuracy within a given training time budget.”
BackgroundInfluentialSemi-Synchronous Personalized Federated Learning Over Mobile Edge Networks
You, C., Feng, D., Guo, K., Yang, H. H., Feng, C., & Quek, T. Q. S. (2022) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2022.3210434 · Cited by 31
“As for the computation time, let ci denote the number of CPU cycles for UE i to execute one sample of data, θi denote the CPU-cycle frequency of UE i, and di denote the number of sampled data points on UE i, then the computation time of UE i per local iteration can be expressed as follows [28],”
MethodEnhancing Federated Learning With Spectrum Allocation Optimization and Device Selection
Zhang, T., Lam, K., Zhao, J., Li, F., Han, H., & Jamil, N. (2023) · IEEE/ACM Transactions on Networking · doi:10.1109/tnet.2022.3231986 · Cited by 36
Clustered Federated Learning in Internet of Things: Convergence Analysis and Resource Optimization
Xu, B., Xia, W., Zhao, H., Zhu, Y., Sun, X., & Quek, T. Q. S. (2023) · IEEE Internet of Things Journal · doi:10.1109/jiot.2023.3298366 · Cited by 36
“1 ms/samples and ς k = ( 1 /χ k ) , respectively [21].”
BackgroundDas, S. K., Mudi, R., Rahman, M. S., Rabie, K. M., & Li, X. (2024) · IEEE Open Journal of Vehicular Technology · doi:10.1109/ojvt.2024.3466858 · Cited by 27
“This approach is particularly beneficial because UEs with superior radio channel conditions can compensate for the interfering signal information from other UEs [210].”
BackgroundMethodDevice Scheduling and Assignment in Hierarchical Federated Learning for Internet of Things
Zhang, T., Lam, K., & Zhao, J. (2024) · IEEE Internet of Things Journal · doi:10.1109/jiot.2024.3362972 · Cited by 37
“At each global iteration, however, [22], [23], [24], [25] require to collect model parameters or loss gradients from all IoT devices to the cloud for making scheduling decisions.”
BackgroundInfluentialCommunication-Efficient Federated Multitask Learning Over Wireless Networks
Ma, H., Guo, H., & Lau, V. K. N. (2022) · IEEE Internet of Things Journal · doi:10.1109/jiot.2022.3201310 · Cited by 32
“3) Best Channel First [37], [38], [39], [40]: In each round, the server schedules one sensor with the largest instantaneous channel power gain.”
BackgroundMethodInfluentialDecentralized Federated Learning With Asynchronous Parameter Sharing for Large-Scale IoT Networks
Xie, H., Xia, M., Wu, P., Wang, S., & Huang, K. (2024) · IEEE Internet of Things Journal · doi:10.1109/jiot.2024.3354869 · Cited by 30
“Specifically, Γ W is from the reserved duration for unscheduled nodes [25], and Γ T is associated with the transmission delay [26].”
Convergence Acceleration in Wireless Federated Learning: A Stackelberg Game Approach
Wang, K., Ma, Y., Mashhadi, M. B., Foh, C. H., Tafazolli, R., & Ding, Z. J. (2024) · IEEE Transactions on Vehicular Technology · doi:10.1109/tvt.2024.3452933 · Cited by 39
“Compared to [18], an energy budget is included and its impact on device selection is analyzed, constructing a more practical and challenging scenario.”
Ma, X., Zhao, J., & Gong, Y. (2021) · IEEE Transactions on Vehicular Technology · doi:10.1109/tvt.2021.3107465 · Cited by 30
“(cid:2) Learning cost: the learning cost is mainly related to communication cost, computation cost and energy consumption [16]–[18], [23].”
BackgroundInfluentialTime-Triggered Federated Learning Over Wireless Networks
Zhou, X., Deng, Y., Xia, H., Wu, S., & Bennis, M. (2022) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2022.3189601 · Cited by 34
“This obtained result is in line with the conclusions of existing works [9], [10], [29].”
MethodResultAdaptive Upgrade of Client Resources for Improving the Quality of Federated Learning Model
AbdulRahman, S., Ould‐Slimane, H., Chowdhury, R., Mourad, A., Talhi, C., & Guizani, M. (2022) · IEEE Internet of Things Journal · doi:10.1109/jiot.2022.3218755 · Cited by 30
“…proposed to select clients based on their resources [8], their communication and computation required time [9], and bandwidth allocation [10], [11], [12], [13], while others developed some techniques which consider the quality of the clients models [14], [15] or when biasing the selection…”
BackgroundDelay-Aware Hierarchical Federated Learning
Lin, F. P.-C., Hosseinalipour, S., Michelusi, N., & Brinton, C. G. (2023) · IEEE Transactions on Cognitive Communications and Networking · doi:10.1109/tccn.2023.3329024 · Cited by 32
“Model Training Under Delay Considerations: Upstream/downstream communication between the edge and server can often lead to non-negligible transmission delays in FL. Delay of model training in FL has been modeled and considered in several recent works [24], [28], [29], [30], [31], [32].”
BackgroundWu, Y., Song, Y., Wang, T., Dai, M., & Quek, T. Q. S. (2022) · IEEE Transactions on Green Communications and Networking · doi:10.1109/tgcn.2022.3164574 · Cited by 27
“To this end, there have been several recent studies exploiting the bandwidth allocation and client-selection in FL for improving the spectrum efficiency [26]–[28].”
BackgroundFedAGL: A Communication-Efficient Federated Vehicular Network
Liu, S., Li, Y., Guan, P., Li, T., Yu, J., Taherkordi, A., & Jensen, C. S. (2024) · IEEE Transactions on Intelligent Vehicles · doi:10.1109/tiv.2024.3349655 · Cited by 28
“Besides the communication overhead, several wireless transmission strategies have been proposed to reduce communication latency [16], [17].”
BackgroundDevice Scheduling in Over-the-Air Federated Learning Via Matching Pursuit
Bereyhi, A., Vagollari, A., Asaad, S., Müller, R. R., Gerstacker, W., & Poor, H. V. (2023) · IEEE Transactions on Signal Processing · doi:10.1109/tsp.2023.3284376 · Cited by 37
“2 The efficiency of this approach has been shown analytically and experimentally through a large set of studies in the literature; see for instance the studies in [30] and [44], [45], [46].”
Efficient Wireless Federated Learning With Partial Model Aggregation
Chen, Z., Yi, W., Shin, H., Nallanathan, A., & Li, G. Y. (2024) · IEEE Transactions on Communications · doi:10.1109/tcomm.2024.3396748 · Cited by 35
Sun, Y., Zhou, S., Niu, Z., & Gündüz, D. (2022) · International Conference on Communications · doi:10.1109/icc45855.2022.9839279 · Cited by 29
“Communication load can also be reduced by enabling multiple iterations of local training, and carefully scheduling a subset of devices in each round [10], [11].”
BackgroundSelf-citationFedPCC: Parallelism of Communication and Computation for Federated Learning in Wireless Networks
Zhang, H., Tian, H., Dong, M., Ota, K., & Jia, J. (2022) · IEEE Transactions on Emerging Topics in Computational Intelligence · doi:10.1109/tetci.2022.3170471 · Cited by 23
“In this subsection, we further conduct experiments to compare FedPCC with baseline FL and FC-like protocol [19] under different learning models, numbers of selected devices, and data distributions.”
BackgroundMethodFederated Learning Over Fully-Decoupled RAN Architecture for Two-Tier Computing Acceleration
Xu, Y., Qian, B., Yu, K. P., Ma, T., Zhao, L. Y., & Zhou, H. (2023) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2023.3236003 · Cited by 28
Low-Latency Hierarchical Federated Learning in Wireless Edge Networks
Su, L., Zhou, R.-T., Wang, N., Chen, J., & Li, Z.-Q. (2024) · IEEE Internet of Things Journal · doi:10.1109/jiot.2023.3314743 · Cited by 23
“1 TABLE II S UMMARY OF N OTATIONS adopted in ML models are linear regression, logistic regression and Neural Network [29].”
MethodSpectrum 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 40
“Diversified approaches have been proposed including radio resource management [9], [10], power con-trol [11], [12], and device scheduling [13], [14].”
BackgroundDynamic Scheduling for Vehicle-to-Vehicle Communications Enhanced Federated Learning
Yan, J., Chen, T., Sun, Y., Nan, Z., Zhou, S., & Niu, Z. (2025) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2025.3573048 · Cited by 29
“It schedules vehicles based on their initial channel states and positions, which is a modified version of the state-of-the-art device scheduling and resource allocation scheme [26].”
Self-citationCommunication-Efficient Federated Learning for Hybrid VLC/RF Indoor Systems
Sheikholeslami, S. M., Rasti-Meymandi, A., Seyedmohammadi, S. J., Abouei, J., & Plataniotis, K. N. (2022) · IEEE Access · doi:10.1109/access.2022.3216710 · Cited by 23
“Authors in [10] proposed a joint device scheduling and resource allocation to maximize the model accuracy in a given training time with latency constraints.”
BackgroundLi, P., Cheng, G., Huang, X., Kang, J., Yu, R., Wu, Y., Pan, M., & Niyato, D. T. (2023) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2023.3245601 · Cited by 27
“In addition, system-level resource management is also an effective technique for energy-efficient FL systems, such as device-to-device communication [15], [16], bandwidth allocation [17], [18], device selection [19], [20] and intelligent reflecting surface technique [21].”
BackgroundMethodEfficient asynchronous federated learning with sparsification and quantization
Jia, J., Liu, J., Zhou, C., Tian, H., Dong, M., & Dou, D. (2024) · Concurrency and Computation Practice and Experience · doi:10.1002/cpe.8002 · Cited by 32
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
“Latency, as one of the important factors in determining the convergence time of FL [8], has been widely studied in existing works [9], [10], [11], [12], [13], [14], [15], [16].”
BackgroundFair Selection of Edge Nodes to Participate in Clustered Federated Multitask Learning
Albaseer, A. M., Abdallah, M. M., Al‐Fuqaha, A., Seid, A. M., Erbad, A., & Dobre, O. A. (2023) · IEEE Transactions on Network and Service Management · doi:10.1109/tnsm.2023.3270168 · Cited by 26
“Focusing on scheduling policies and participants’ selection, to minimize the latency during the FL training process, several client scheduling techniques were proposed in [1], [25], [26], [32], [33], [34], [35].”
BackgroundInfluentialCan Decentralized Control Outperform Centralized? The Role of Communication Latency
Ballotta, L., Jovanović, M. R., & Luca Schenato, B. (2023) · IEEE Transactions on Control of Network Systems · doi:10.1109/tcns.2023.3237483 · Cited by 20
“…low-power devices (e.g., TinyML [6]), which allow to spread computational tasks across network nodes according to edge-and fog-computing paradigms [7], [8], [9], are making such networked systems Luca Ballotta and Luca Schenato are with the Department of Information Engineering, University of…”
BackgroundClient Selection with Bandwidth Allocation in Federated Learning
Kuang, J., Yang, M., Zhu, H., & Qian, H. L. (2021) · Globecom · doi:10.1109/globecom46510.2021.9685090 · Cited by 24
“To simplify the problem, we neglect the time consumption by broadcasting and aggregation [7].”
MethodMultiple Parallel Federated Learning via Over-the-Air Computation
Shi, G., Guo, S., Ye, J., Saeed, N., & Dang, S. (2022) · IEEE Open Journal of the Communications Society · doi:10.1109/ojcoms.2022.3194821 · Cited by 24
“Remark 1: Assumptions 1-4 are satisfied for most machine learning loss functions, such as squared support vector machine (SVM) and linear regression [41], and are widely used in the literature of performance analysis for FL [10], [42].”
MethodLatency-Efficient Wireless Federated Learning With Quantization and Scheduling
Yan, Z., Li, D., Yu, X., & Zhang, Z. (2022) · IEEE Communications Letters · doi:10.1109/lcomm.2022.3199490 · Cited by 23
“Although there have been already some works studying the latency-efficient FL [6], [7], [8], [9], [10], [11], the impact of joint quantization and scheduling and how to obtain the optimized quantization level and scheduling number is still unknown, which motivates this work.”
BackgroundMethodInfluentialEdge Learning for Large-Scale Internet of Things With Task-Oriented Efficient Communication
Xie, H., Xia, M., Wu, P., Wang, S., & Poor, H. V. (2023) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2023.3271665 · Cited by 24
“In particular, as the number of scheduling cases increases exponentially with the number of users, it is hard to enumerate all possible subsets of users explicitly [20].”
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].”
MethodZhao, H., Tan, Y., Guo, K., Xia, W., Xu, B., & Quek, T. Q. S. (2024) · IEEE Internet of Things Journal · doi:10.1109/jiot.2024.3354914 · Cited by 24
“For instance, Shi et al. [19] studied the joint client scheduling and resource allocation problem of latency constrained wireless FL, which aims to maximize the global model accuracy under given total training time constraint.”
BackgroundComputation Offloading for Edge Intelligence in Two-Tier Heterogeneous Networks
Zhao, J., Li, Q., Ma, X., & Yu, F. R. (2023) · IEEE Transactions on Network Science and Engineering · doi:10.1109/tnse.2023.3332949 · Cited by 24
“In [20], [21], the authors exploited a joint device scheduling and resource allocation policy to maximize the learning accuracy for wireless federated learning.”
BackgroundFederated Learning-Based Resource Allocation for V2X Communications
Bhardwaj, S., Kim, D.-H., & Kim, D. (2024) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2024.3500004 · Cited by 26
Le, T. H. T., Cantos, L., Pandey, S. R., Shin, H., & Kim, Y. H. (2023) · IEEE Transactions on Vehicular Technology · doi:10.1109/tvt.2023.3264202 · Cited by 21
Truthful Incentive Mechanism for Federated Learning with Crowdsourced Data Labeling
Zhao, Y., Gong, X., & Mao, S. (2023) · IEEE Conference on Computer Communications (INFOCOM) · doi:10.1109/infocom53939.2023.10228923 · Cited by 22
Coalitional FL: Coalition Formation and Selection in Federated Learning With Heterogeneous Data
Zhang, N., Ma, Q., Mao, W., & Chen, X. (2024) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2024.3375325 · Cited by 23
“6 Here we assume that the server only cares about the performance (i.e., the model accuracy) of FL and has a large enough budget to pay rewards to devices [36].”
BackgroundTang, J., Li, X., Li, H., Li, P., Wang, X., & Leung, V. C. M. (2025) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2025.3539284 · Cited by 23
“Besides, its bandwidth allocation only considered the latency, just like [16], [17], but the former is a simplified version of the latter.”
InfluentialFedSeq: A Hybrid Federated Learning Framework Based on Sequential In-Cluster Training
Chen, Z., Li, D., Ni, R., Zhu, J., & Zhang, S. (2023) · IEEE Systems Journal · doi:10.1109/jsyst.2023.3243694 · Cited by 22
“For the computation latency at an arbitrary round, we use the shifted exponential distribution [33] to characterize the computation latency of local model updating of user i:”
MethodThe Node Selection Strategy for Federated Learning in UAV-Assisted Edge Computing Environment
Bai, J., & Chen, Y. (2023) · IEEE Internet of Things Journal · doi:10.1109/jiot.2023.3263359 · Cited by 18
“Some widely used loss functions, for example, logistic regression, mean-squared error, and cross entropy, satisfy these assumptions [20], [21].”
MethodGoal-Oriented Wireless Communication Resource Allocation for Cyber-Physical Systems
Feng, C., Zheng, K., Wang, Y., Huang, K., & Chen, Q. (2024) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2024.3432918 · Cited by 25
“[33]– [35] employed various criteria to identify the importance of gradient updates, subsequently scheduling participation and consequently reducing the overall training time.”
BackgroundDistributed Learning in Intelligent Transportation Systems: A Survey
Li, Q., Zhou, W., & Zheng, X. (2024) · Information · doi:10.3390/info15090550 · Cited by 24
Adaptive Training and Aggregation for Federated Learning in Multi-Tier Computing Networks
Hou, W., Wen, H., Zhang, N., Lei, W., Lin, H., Han, Z., Liu, Q., & Tian, W. (2023) · Elsevier eBooks · doi:10.1016/b978-0-44-344433-3.00020-4 · Cited by 20
Singh, A. K., & Nguyen, K. K. (2024) · IEEE/ACM Transactions on Networking · doi:10.1109/tnet.2024.3384839 · Cited by 31
“In prior works [15], [16], [17], [18], and [19], the authors have tackled another aspect of FL performance.”
BackgroundChang, Z.-L., Hosseinalipour, S., Chiang, M., & Brinton, C. G. (2024) · IEEE Transactions on Cognitive Communications and Networking · doi:10.1109/tccn.2024.3391329 · Cited by 25
“Reference [18] proposed a joint device scheduling and resource allocation policy to maximize the model accuracy within a given total training time budget for latency constrained wireless FL. Reference [19] proposed data-aware wireless scheduling algo-rithm to minimize the completion time and the…”
BackgroundFedRelay: Federated Relay Learning for 6G Mobile Edge Intelligence
Li, P., Zhong, Y., Zhang, C., Wu, Y., & Yu, R. (2022) · IEEE Transactions on Vehicular Technology · doi:10.1109/tvt.2022.3225087 · Cited by 21
“…j in flow n , the achievable data transmission rate r n,i is calculated by where | h i → j | is the channel power gain between device i and its preceding relay node; b n is the available bandwidth of flow n , and P com t,i is the transmission power of device i ; σ 2 is the noise power density [35].”
BackgroundResultOptimal Mechanism Design for Heterogeneous Client Sampling in Federated Learning
Liao, G., Luo, B., Feng, Y., Zhang, M., & Chen, X. (2024) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2024.3379659 · Cited by 19
“Many studies focused on optimizing FL systems from different technical perspectives such as model accuracy (e.g., [20], [21], [22]), implementation costs (e.g., [5], [10], [23], [24], [25]), and convergence rates (e.g., [26]).”
BackgroundStragglers-Aware Low-Latency Synchronous Federated Learning via Layer-Wise Model Updates
Lang, N., Cohen, A., & Shlezinger, N. (2024) · IEEE Transactions on Communications · doi:10.1109/tcomm.2024.3486979 · Cited by 25
“Such applications include, e.g., intelligent transportation systems [7], and wireless networks adaptation [8], [9], [10].”
Data Partition and Rate Control for Learning and Energy Efficient Edge Intelligence
Li, X., Wang, S., Zhu, G., Zhou, Z., Huang, K., & Gong, Y. (2022) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2022.3173262 · Cited by 16
Federated learning in healthcare: Recent progress and challenges
Miloudi, A., Laouid, A., Bounceur, A., Kara, M., Bouhamed, M. M., Hammoudeh, M., & Kraidia, I. (2026) · Computers & Electrical Engineering · doi:10.1016/j.compeleceng.2025.110924 · Cited by 19
Chen, X.-J., Li, Z.-Y., Ni, W., Wang, X., Zhang, S.-Q., Xu, S.-G., & Pei, Q.-Q. (2022) · 2022 IEEE/CIC International Conference on Communications in China (ICCC) · doi:10.1109/iccc55456.2022.9880724 · Cited by 10
“On the other hand, it has been demonstrated from empirical studies [13], [14] and theoretical analysis [8] that more participating clients in an FL process can improve the convergence rate and model accuracy.”
BackgroundQ-Learning-Aided Offloading Strategy in Edge-Assisted Federated Learning over Industrial IoT
Wu, S., Xue, H., & Zhang, L. (2023) · Electronics · doi:10.3390/electronics12071706 · Cited by 15
Yan, S., Zhang, P., Huang, S., Wang, J., Sun, H., Zhang, Y., & Tolba, A. K. R. (2023) · Electronics · doi:10.3390/electronics12112478 · Cited by 20
“[24] proposed a joint equipment scheduling and resource allocation strategy.”
BackgroundEfficient Federated Learning via Joint Communication and Computation Optimization
Wang, G., Zhao, C., Qi, Q., Han, R., Bai, L., & Choi, J. (2024) · IEEE Transactions on Vehicular Technology · doi:10.1109/tvt.2024.3379742 · Cited by 16
“The joint optimization of scheduling policy and bandwidth allocation strategy was studied in [8], [9].”
BackgroundA Comprehensive Review on Machine Learning-based Approaches for Next Generation Wireless Network
Paul, S. (2024) · SN Computer Science · doi:10.1007/s42979-024-02831-3 · Cited by 17
“Furthermore, applying FL for the scheduling of a small part of devices at every turn is an alternate promising approach [166, 167] for network performance optimization.”
Rethinking Clustered Federated Learning in NOMA Enhanced Wireless Networks
Lin, Y., Wang, K., & Ding, Z. (2024) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2024.3447833 · Cited by 16
“In [22], a study of maximizing model accuracy within a given total training time budget in latency-constrained FL in wireless networks is investigated.”
BackgroundCF4FL: A Communication Framework for Federated Learning in Transportation Systems
Sangdeh, P. K., Li, C., Pirayesh, H., Zhang, S., Zeng, H., & Hou, Y. T. (2022) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2022.3221770 · Cited by 18
“Several research efforts have been made to study participant scheduling and resource allocation toward different objectives, such as minimizing FL loss [10]–[12], minimizing latency [13], improving energy efficiency [14]–[16], and enhancing learning efficiency [28].”
BackgroundLee, H.-S., & Lee, D. (2022) · ICT express · doi:10.1016/j.icte.2022.01.019 · Cited by 17
Sun, H., Zhang, H., Ma, H., & Leung, V. C. M. (2025) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2025.3567289 · Cited by 13
“In [26], W. Shi et al. proposed a FRL-based device scheduling and communication resource allocation strategy in large-scale cellular networks with massive distributed network data.”
Xu, Z., Wang, L., Liang, W., Xia, Q., Xu, W., Zhou, P., & RANA, O. F. (2023) · IEEE Transactions on Computers · doi:10.1109/tc.2023.3333213 · Cited by 19
“Although the communication delay is a vital part of AoD, existing studies on optimizing communication delay for federated learning cannot be directly applied to FCL in MEC [20], [32], [33], [39], [41], [42].”
BackgroundBandwidth Allocation for Federated Learning With Wireless Providers and Cost Constraints
Xue, J., Qu, Z., Xu, J., Liu, Y., & Lu, Z. (2023) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2023.3335258 · Cited by 15
“Bandwidth Allocation: Bandwidth allocation approaches have been proposed to improve the service efficiency over wireless networking [3], [4], [9], [13], [14], [20], [44], [47], [48], [49], [50], [51], [52].”
BackgroundInfluentialLong-Term Energy Management Empowered Hierarchical Federated Learning for Smart Consumer Electronics
Xu, B., Zhao, H., Cao, H., Garg, S., & Alrashoud, M. R. (2024) · IEEE Transactions on Consumer Electronics · doi:10.1109/tce.2024.3355058 · Cited by 16
“The local transmit power of devices is p max [32], the local computing power budget of all devices is f max 5 GHz, and the energy budget of all devices is ¯ E 1 = · · · = ¯ E K = 0 .”
BackgroundJoint Client Selection and Resource Allocation for Federated Learning in Mobile Edge Networks
Luo, L., Cai, Q., Li, Z., & Yu, H. (2022) · IEEE Wireless Communications and Networking Conference · doi:10.1109/wcnc51071.2022.9771771 · Cited by 13
“Yang et al. , [8], Xu et al. , [5] and Shi et al. , [9] focus on the allocation of bandwidth and channel resource for achieving long-term performance guarantee, minimizing FL delay, and maximizing model accuracy within a given total learning time budget, respectively.”
BackgroundZhao, H., Geng, L., Feng, W., & Changming, Z. H. O. U. (2024) · Chinese Journal of Aeronautics · doi:10.1016/j.cja.2024.06.023 · Cited by 13
Self-citationFederated learning for digital healthcare: concepts, applications, frameworks, and challenges
Sachin, D. N., Annappa, B., & Ambesange, S. (2024) · Computing · doi:10.1007/s00607-024-01317-7 · Cited by 11
“Shi et al. [61] investigates another combined optimization method for resource allocation and device scheduling.”
Device Selection and Resource Allocation for Layerwise Federated Learning in Wireless Networks
Lee, H.-S. (2022) · IEEE Systems Journal · doi:10.1109/jsyst.2022.3169461 · Cited by 13
“A variety of works have been widely studied on DSRA for FL in wireless networks with the consideration of the characteristics of wireless networks [8]–[12].”
BackgroundWu, Z., Wu, X., & Long, Y. (2022) · IEEE Transactions on Network and Service Management · doi:10.1109/tnsm.2022.3232381 · Cited by 12
“Therefore, the achievable rate of client i receiving packets from client j can be denoted as Thus the transmission latency caused by the information transmitted from client j to client i can be expressed as where S represents the size of model parameters, in bits [37].”
BackgroundCFLIT: Coexisting Federated Learning and Information Transfer
Lin, Z., Liu, H., & Zhang, Y. A. (2023) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2023.3263148 · Cited by 14
“In this paper, we assume perfect downlink model broad-casting following [10], [11], [12], [13], [24], [34] and focus on the uplink communication design for over-the-air model aggregation, which is elaborated as follows.”
MethodOn the Optimization of UAV-Assisted Wireless Networks for Hierarchical Federated Learning
Khelf, R., Driouch, E., & Ajib, W. (2023) · IEEE International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC) · doi:10.1109/pimrc56721.2023.10293857 · Cited by 13
“Studies in [10] and [11] have demonstrated that increasing FL users improve model accuracy and convergence rate.”
BackgroundWang, T., Huang, N., Wu, Y., Gao, J., & Quek, T. Q. S. (2023) · IEEE Internet of Things Journal · doi:10.1109/jiot.2023.3234422 · Cited by 12
“…[34] proposed a heterogeneous computation and resource allocation framework based on a heterogeneous mobile architecture to achieve an effective implementation of FL. Shi et al. [35] improved the model accuracy of wireless FL by jointly optimizing the device scheduling and the resource allocation.”
MethodLatency Minimization for Wireless Federated Learning With Heterogeneous Local Model Updates
Zhu, J., Shi, Y., Fu, M., Zhou, Y., Wu, Y., & Fu, L. (2023) · IEEE Internet of Things Journal · doi:10.1109/jiot.2023.3285937 · Cited by 12
“Shi et al. [41] proposed a greedy device selection algorithm based on the minimum single-round latency.”
BackgroundMethodResultInfluentialA Flexible Model Compression and Resource Allocation Scheme for Federated Learning
Hu, Y., Liu, T., Yang, C., Huang, Y., & Suo, S. (2023) · IEEE Transactions on Machine Learning in Communications and Networking · doi:10.1109/tmlcn.2023.3303292 · Cited by 10
“To demonstrate the advantages of the proposed method, we need to compare it with existing joint optimization approaches [3], [11], [15], [16].”
BackgroundMethodInfluentialMobility-Aware Joint User Scheduling and Resource Allocation for Low Latency Federated Learning
Fan, K., Chen, W., Li, J., Deng, X., Han, X., & Ding, M. (2023) · IEEE/CIC International Conference on Communications in China (ICCC) · doi:10.1109/iccc57788.2023.10233347 · Cited by 12
“Considering that the power and bandwidth of the BS are sufficient, the download latency is negligible compared with the total delay [9].”
BackgroundInfluentialOptimizing Tradeoff Between Learning Speed and Cost for Federated-Learning-Enabled Industrial IoT
Zhang, L., Wu, S., Xu, H., Liu, Q., Hong, C. S., & Han, Z. (2023) · IEEE Internet of Things Journal · doi:10.1109/jiot.2023.3330754 · Cited by 12
“Different from the above solutions [9], [15], [16] that considered only a single training round, Xu and Wang [17] formulated a stochastic optimization problem for joint client selection and bandwidth allocation under the long-term client energy constraint to improve the learning performance.”
BackgroundMethodKeçeci, C., Shaqfeh, M. O., Al‐Qahtani, F. S., Ismail, M., & Serpedin, E. (2023) · IEEE Internet of Things Journal · doi:10.1109/jiot.2023.3262620 · Cited by 12
Yu, C., Meng, Z., Zhang, W., Lei, L., Ni, J., Zhang, K., & Hai, Z. (2024) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2024.3486028 · Cited by 14
“According to [34], [35], [36], the achievable transmission rate (bits/s) for the uplink between device m and the cloud server is where P m represents the transmission power of device m , which remains constant.”
Energy-Efficient Federated Edge Learning in Multi-Tier NOMA-Enabled HetNet
Hossain, M. A., & Ansari, N. (2023) · IEEE Transactions on Cloud Computing · doi:10.1109/tcc.2023.3285534 · Cited by 9
“…in wireless networks such as wireless resource allocation [15], learning rate optimization [16], algorithm and convergence analysis [17], spatial convergence analysis [18], incentive analysis [19], and device scheduling approaches [20], [21], [22] have been studied widely over the last few years.”
BackgroundHong, J., Park, S., & Choi, W. (2023) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2023.3249252 · Cited by 13
“In particular, edge device scheduling schemes have been proposed to increase training efficiency by solving the related optimization problems [11], [12], [13], [14], [15].”
BackgroundLi, Y., Qin, X., Geng, J., Chen, R., Hou, Y., Gong, Y., Pan, M., & Zhang, P. (2024) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2024.3365477 · Cited by 9
“In [31], the authors allocated more bandwidth to participants with limited computational resources to decrease training latency.”
BackgroundFedCarbon: Carbon-Efficient Federated Learning with Double Flexible Controls for Green Edge AI
Li, Y., Ouyang, T., Chen, X., & Cao, X. (2024) · International Workshop on Quality of Service · doi:10.1109/iwqos61813.2024.10682880 · Cited by 16
“To alleviate such performance deteriorations, some works [5], [26] have devised client selection strategies with rapid convergence in the heterogeneous resources environment, but non-IID data distribution severely impacts the model accuracy.”
BackgroundDevice Selection and Resource Allocation With Semi-Supervised Method for Federated Edge Learning
Hu, R., Yuan, H., Tan, D., & Wang, Z. (2024) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2024.3504271 · Cited by 8
“Despite these constraints, FEL offers numerous opportunities alongside challenges, encompassing multiple access scenarios [26], [27], [28], task scheduling [29], [30], network bandwidth [14], [15], [16], [17], and cache resource management [18], [19].”
InfluentialLi, P., Dong, H., Qian, L., Zhou, S., & Wu, Y. (2024) · IEEE Transactions on Cognitive Communications and Networking · doi:10.1109/tccn.2024.3522084 · Cited by 10
“Similar to several previous studies on deep learning analysis [25], [37], we make the following two assumptions for the single-step inference of the diffusion model.”
Self-citationWang, Z., Shu, Y., Liao, H., Zhou, Z., Lv, L., Zhang, L., Mumtaz, S., & Guizani, M. (2025) · IEEE Transactions on Smart Grid · doi:10.1109/tsg.2025.3526437 · Cited by 11
“The fast converge scheduling-based resource management strategy (FCS-RMS) [13] and device scheduling-based resource management strategy (DS-RMS) [23] for FL-based model training are used as comparison algorithms.”
Federated Low-Rank Adaptation for Large Language Model Fine-Tuning Over Wireless Networks
Wang, Z., Zhou, Y., Shi, Y.-M., & Letaief, K. (2024) · Global Communications Conference · doi:10.1109/globecom52923.2024.10901572 · Cited by 8
“To facilitate an efficient high-dimensional model exchange under limited radio resources, many of existing approaches have been investigated from the perspective of device scheduling [12], power allocation [13], beamforming design [14], and bandwidth optimization [15].”
PairingFL: Efficient Federated Learning With Model Splitting and Client Pairing
Yao, Z., Qi, J., Xu, Y., Liao, Y., Xu, H., & Wang, L. (2025) · IEEE Transactions on Networking · doi:10.1109/ton.2025.3547984 · Cited by 15
“In addition, some other works [17], [32], [33], [34], [35] focus on overcoming challenges related to system heterogeneity but eventually experience accuracy degradation when dealing with non-IID data.”
Breakout Local Search for Load-Balanced Federated Learning in Multi-BS Networks
Kato, M., Yuan, X., Rodrigues, T. K., Tang, F., Zhao, M., & Kato, N. (2026) · IEEE Transactions on Emerging Topics in Computing · doi:10.1109/tetc.2026.3667101 · Cited by 9
“The over-all round time, however, is gated by the slowest selected participant [4].”
A General Solution for Straggler Effect and Unreliable Communication in Federated Learning
Zang, T., Zheng, C., Ma, S., Sun, C., & Chen, W. (2023) · International Conference on Communications · doi:10.1109/icc45041.2023.10279635 · Cited by 17
“We refer to and use most of the setup in [17].”
MethodDPBalance: Efficient and Fair Privacy Budget Scheduling for Federated Learning as a Service
Liu, Y., Wang, Z., Zhu, Y., & Chen, C. (2024) · IEEE Conference on Computer Communications (INFOCOM) · doi:10.1109/infocom52122.2024.10621227 · Cited by 12
“In [20], a joint device scheduling and resource allocation policy is proposed to maximize model accuracy with a limited training time budget and latency threshold.”
BackgroundClient Scheduling for Federated Learning over Wireless Networks: A Submodular Optimization Approach
Ye, L., & Gupta, V. (2021) · IEEE Conference on Decision and Control (CDC) · doi:10.1109/cdc45484.2021.9683685 · Cited by 12
“The FL client scheduling problem has been studied under various different scenarios (e.g., [8]–[10]).”
BackgroundMethodInfluentialHe, S., Ren, T., Jiang, X.-H., & Xu, M. (2023) · IEEE Wireless Communications and Networking Conference · doi:10.1109/wcnc55385.2023.10118927 · Cited by 7
Bi, M., Yan, C., Lin, R., Hong, Y., & Li, J. (2024) · Computer Networks · doi:10.1016/j.comnet.2024.110309 · Cited by 7
Meng, Z., Li, Z. Z., Hou, X., Xu, M., Xia, Y., Zhang, Z., & Song, S. (2024) · IEEE Internet of Things Journal · doi:10.1109/jiot.2024.3521017 · Cited by 9
“Based on the above three assumptions [34], [35], [36], we can calculate the lower bound of the number of local training sessions [37].”
Energy Efficient and Low Latency Federated Distillation Over UAV-Assisted Wireless Networks
Zhang, Z., Zhao, Y., Chen, C., Zhu, K., & Niyato, D. T. (2025) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2025.3557832 · Cited by 12
“According to [28], when the location of the air parameter server UAV is fixed, if and only if all the bandwidth resources of the device are allocated and have approximately the same completion time, i.e., solve the following linear equation in one variable: Therefore, the optimal bandwidth…”
HASFL: Heterogeneity-Aware Split Federated Learning Over Edge Computing Systems
Lin, Z., Chen, Z., Chen, X., Ni, W., & Gao, Y. (2026) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2026.3673358 · Cited by 56
“For simplicity, the sub-model aggregation latency for this part is ignored, as it is negligible compared to the latency from other stages [42], [43]. c) Sub-model downloading latency: After completing the client-side model aggregation, the fed server sends the updated client-side sub-models to the…”
Incentive-driven long-term optimization for hierarchical federated learning
Su, L., & Li, Z. (2023) · Computer Networks · doi:10.1016/j.comnet.2023.109944 · Cited by 7
Joint 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
“…training, the gap to the global optimal loss function is analyzed versus the number of iterations, which is expected to converge to zero with a high-convergence rate, under various configurations of step size and searching gradient [20], [21], [22], [26], [27], [28], [29], [30], [31], [32], [33].”
Background
Metadata
- Authors
- Shi, W., Zhou, S., Niu, Z., Jiang, M., & Geng, L.
- Affiliations
- Beijing National Research Center for Information Science and Technology (CN)
- Hitachi (Japan) (JP)
- Venue
- IEEE Transactions on Wireless Communications
- Published
- 2021-01
- Volume
- 20
- Issue
- 1
- Pages
- 453-467
- ISSN
- 1536-1276 (ISSN-L)
- 1558-2248
- OpenAlex
- W3089655738
- Cited by
- 381
- References
- 33
- Citations by year
- 2026: 32 · 2025: 48 · 2024: 80 · 2023: 106 · 2022: 75 · 2021: 38
- FWCI
- 21.70
- 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 Key Research and Development Program of China (2018YFB0105000)
- Nature Science Foundation of China (61871254, 91638204, 61861136003)
- Hitachi Ltd
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Privacy-Preserving Technologies in Data
- Physical Sciences › Engineering › Electrical and Electronic Engineering › Advanced Wireless Network Optimization
- Physical Sciences › Computer Science › Computer Networks and Communications › IoT and Edge/Fog Computing
- Keywords
- wireless federated learning, device scheduling, resource allocation, bandwidth allocation, delay constraints, lower bounds, greedy scheduling algorithm, convergence rate, channel conditions
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.