Federated Learning for Edge Networks: Resource Optimization and Incentive Mechanism
Open accessAbstract
Recent years have witnessed a rapid proliferation of smart Internet of Things (IoT) devices. IoT devices with intelligence require the use of effective machine learning paradigms. Federated learning can be a promising solution for enabling IoT-based smart applications. In this article, we present the primary design aspects for enabling federated learning at the network edge. We model the incentive- based interaction between a global server and participating devices for federated learning via a Stackelberg game to motivate the participation of the devices in the federated learning process. We present several open research challenges with their possible solutions. Finally, we provide an outlook on future research.
AI snapshot
Written from the paper’s full text.
TL;DR
This article presents design aspects for federated learning at the network edge and models the interaction between a base station and participating devices as a Stackelberg game to incentivize participation. It evaluates an incentive-based FL model using MNIST distributed across five UEs and outlines open challenges in resource optimization, context awareness, and mobility.
Key findings
- The paper proposes a Stackelberg game with one leader (the BS) and multiple followers (UEs), where UEs strategically choose local iteration/CPU-frequency strategies and the BS adjusts reward rates to maximize FL performance.
- It identifies four key design aspects for edge FL: resource optimization, incentive mechanism design, learning algorithm design, and hardware-software co-design.
- In the performance evaluation, the offered reward rate motivates UEs to perform more local iterations within one global iteration, improving relative accuracy; heterogeneous UE responses arise from individual computational limitations, local data size, and communication channel conditions.
- The evaluation uses a classification task with multinomial logistic regression, distributes the MNIST dataset among five participating UEs with equal local data size, and uses the modified CoCoA framework for federated optimization.
Methods
- Design: conceptual/overview article with a Stackelberg game-based incentive mechanism for FL over edge networks.
- Data/sample: simulation-style evaluation with five participating UEs having different channel conditions and equal local data size; MNIST dataset distributed among UEs; classification via multinomial logistic regression; modified CoCoA for federated optimization.
- Analysis: best-response algorithm to reach Stackelberg equilibrium; first-order condition to find a unique Nash equilibrium among UEs at the lower level, then backward induction to solve the upper-level BS problem; performance shown through figures on reward rate, relative accuracy, communication time, and normalized BS utility.
Limitations
- The paper does not state a dedicated limitations section.
- Evident from the text: the performance evaluation is small-scale, using only five UEs with equal local data size on MNIST, so the system model’s heterogeneous data sizes are not tested in the reported evaluation.
- Evident from the text: no numerical performance values are reported in the text; results are described qualitatively through figure axes and trends.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
0 of 17 cited sources checked · 0 supported · 0 flagged
- Findingnone given
The offered reward rate reveals the optimal response of the UEs that maximizes their own utilities for given channel conditions.
“The offered reward rate reveals the optimal response of the UEs that maximizes their own utilities for given channel conditions.”
Quote in the paperSubstantial claim - Findingnone given
There is consistency in the normalized BS utility function for various response behaviors of the UEs to the offered reward rate.
“Here, we have consistency in the normalized BS utility function for various response behaviors of the UEs to the offered reward rate.”
Quote in the paperNot a substantial claim - Findingreasoning
In Fig. 3a, the accuracy improves when the relative accuracy value (x-axis) is smaller.
“The accuracy improves when the relative accuracy value (x-axis) is smaller.”
Quote in the paperNot a substantial claim - Findingreasoning
As the communication time increases, the UEs perform more local iterations to avoid expensive communication costs.
“As the communication time increases, the UEs perform more local iterations to avoid expensive communication costs.”
Quote in the paperSubstantial claim - Interpretationnone given
The heterogeneous UE responses is the result of individual computational limitations, local data size, and communication channel conditions.
“The heterogeneous UE responses is the result of individual computational limitations, local data size, and communication channel conditions.”
Quote in the paperSubstantial claim
- Interpretationreasoning
With the first-order condition, there is a unique Nash equilibrium at the lower-level problem (among UEs).
“For this, with the first-order condition, we first find a unique Nash equilibrium at the lower-level problem (among UEs), and, then, use a backward induction method to solve the upper-level BS problem.”
Quote in the paperSubstantial claim- The authors state this as an assumption: The utility of the BS is modeled as a strictly concave function of key FL performance metrics such as the number of global iterations required to reach global accuracy for a given local relative accuracy.
- The authors state this as an assumption: For the UEs utility, we choose a concave function of the local relative accuracy and the BS-offered reward.
- Premisereasoning
By doing so, FedProx limits the impact of non-independent and identically distributed (non-IID) device data on the global learning model.
“By doing so, FedProx limits the impact of non-independent and identically distributed (non-IID) device data on the global learning model.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [7] source
Not checked in the free run.
- Premisecitation
To introduce potential fairness and reduce training accuracy variance, local devices having a high empirical loss are emphasized by assigning higher relative weight in the fairness enabled FedAvg.
“To introduce potential fairness and reduce training accuracy variance, local devices having a high empirical loss (local loss function) are emphasized by assigning higher relative weight in the fairness enabled FedAvg.”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [8] source
Not checked in the free run.
- Interpretationreasoning
The complex interaction between the BS and the UEs can be cast as a Stackelberg game with one leader (BS) and multiple followers (UEs).
“This complex interaction between the BS and the UEs can be cast as a Stackelberg game with one leader (BS) and multiple followers (UEs).”
Quote in the paperSubstantial claim- Rests on an assumption: Heterogeneous UEs have different computational and communication costs for training and, thus, they expect different rewards.
- Premisecitation
Selecting the IoT devices with sufficient computing power and training data jointly improves FL model accuracy and training costs.
“Selecting the IoT devices with sufficient computing power and training data, jointly improves FL model accuracy and training costs [4].”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [4]
Not checked in the free run.
- Premisecitation
The distributed federated optimization scheme of [10] allows us to tackle both system-level and statistical heterogeneity efficiently.
“The distributed federated optimization scheme of [10] allows us to tackle both system-level and statistical heterogeneity efficiently.”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [10]
Not checked in the free run.
- Interpretationreasoning
Game-theoretic incentive mechanisms can be used to effectively model interaction between devices and edge server for FL.
“We have shown that game-theoretic incentive mechanisms can be used to effectively model interaction between devices and edge server for FL.”
Quote in the paperSubstantial claim- Rests on an assumption: UEs are generally reluctant to participate in FL due to limited computing and communication resources.
- The authors state this as an assumption: The utility of the BS is modeled as a strictly concave function of key FL performance metrics such as the number of global iterations required to reach global accuracy for a given local relative accuracy.
- The authors state this as an assumption: For the UEs utility, we choose a concave function of the local relative accuracy and the BS-offered reward.
- Rests on an assumption: Heterogeneous UEs have different computational and communication costs for training and, thus, they expect different rewards.
- Premisecitation
For synchronous aggregation, given a fixed number of global FL rounds between end-devices and edge server, the convergence rate of the global FL model has a proportional relationship with the number of local iterations.
“For synchronous aggregation, given a fixed number of global FL rounds between end-devices and edge server, the convergence rate of the global FL model has a proportional relationship with the number of local iterations.”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [4]
Not checked in the free run.
- Classified as probably not a substantial claim. 9
- May combine two claims. 6
Classified as probably not a substantial claim.
Findingnone givenAn increase in the offered reward rate will motivate UEs to iterate more within one global iteration, resulting in a better accuracy.
“Intuitively, an increase in the offered reward rate will motivate UEs to iterate more within one global iteration, resulting in a better accuracy.”
Quote in the paperSubstantial claimMay combine two claims.
Premisenone givenWorks [2] and [3] rely on centralized learning solutions that have limited scalability and privacy-preservation.
“Works in [2] and [3] rely on centralized learning solutions that have limited scalability and privacy-preservation.”
Quote in the paperSubstantial claim0 of 2 sources checked- cites [2]
Not checked in the free run.
- cites [3] source
Not checked in the free run.
Probably not a substantial claim: may not pass S4 (based).
Premisenone givenWorks [5] and [6] do not discuss the important challenges pertaining to incentive design and network optimization under edge-based FL.
“However, the works in [5] and [6] do not discuss the important challenges pertaining to incentive design and network optimization under edge-based FL.”
Quote in the paperNot a substantial claim0 of 2 sources checked- cites [5] source
Not checked in the free run.
- cites [6]
Not checked in the free run.
May combine two claims.
PremisecitationFedAvg does not guarantee theoretical convergence, while FedProx shows theoretical convergence.
“FedAvg does not guarantee theoretical convergence, while FedProx shows theoretical convergence.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [7] source
Not checked in the free run.
Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing), S4 (based).
InterpretationreasoningThis work can potentially make FL amenable for implementation in diverse 5G-enabled smart IoT applications such as intelligent transportation systems, industry 4.0, and digital health care.
“This work can potentially make FL amenable for implementation in diverse 5G-enabled smart IoT applications such as intelligent transportation systems, industry 4.0, and digital health care.”
Quote in the paperNot a substantial claim- Rests on an assumption: Federated learning can be a promising solution for enabling IoT-based smart applications.
Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.
PremisecitationThe problem studied in [15] captures two tradeoffs: (a) UE energy consumption and FL time via variations in device CPU-cycles/sec and (b) computational and communication latencies for FL accuracy.
“The problem studied in [15] captures two tradeoffs: (a) UE energy consumption and FL time via variations in device CPU-cycles/sec and (b) computational and communication latencies for FL accuracy.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [15] source
Not checked in the free run.
May combine two claims.
InterpretationreasoningAn increase in the number of local iterations minimizes the local learning model error and thus, few global FL rounds are required to reach a certain global FL model accuracy.
“An increase in the number of local iterations minimizes the local learning model error and thus, few global FL rounds are required to reach a certain global FL model accuracy.”
Quote in the paperSubstantial claimProbably not a substantial claim: may not pass S3 (load-bearing).
PremisecitationIn our previous work [15], we jointly optimized the computing time and energy consumption of FL over wireless networks.
“In our previous work [15], we jointly optimized the computing time and energy consumption of FL over wireless networks.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [15] source
Not checked in the free run.
Probably not a substantial claim: may not pass S3 (load-bearing).
PremisecitationA federated multitask learning scheme was proposed in [10] by modifying the CoCoA framework.
“A federated multitask learning scheme was proposed in [10] by modifying the so-called communication-efficient distributed dual coordinate ascent (CoCoA) framework.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [10]
Not checked in the free run.
May combine two claims.
PremisecitationThe UEs will respond to the BS-offered reward and choose their local iteration strategy to maximize their own benefits.
“Correspondingly, the UEs will respond to the BS-offered reward and choose their local iteration strategy (i.e., select a CPU-frequency cycle for local computation) to maximize their own benefits [14].”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [14]
Not checked in the free run.
- Rests on an assumption: Heterogeneous UEs have different computational and communication costs for training and, thus, they expect different rewards.
Probably not a substantial claim: may not pass S3 (load-bearing).
PremisecitationTo enable a wide variety of machine learning models, CoCoA supports objectives for linear regularized loss minimization.
“To enable a wide variety of machine learning models, CoCoA supports objectives for linear reguarlized loss minimization [11].”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [11]
Not checked in the free run.
Probably not a substantial claim: may not pass S3 (load-bearing).
PremisecitationThis adaptive control scheme offers a desirable tradeoff between global model aggregation and local model update to minimize the loss function with resource budget constraint.
“This adaptive control scheme offers a desirable tradeoff between global model aggregation and local model update to minimize the loss function with resource budget constraint.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [9] source
Not checked in the free run.
Probably not a substantial claim: may not pass S3 (load-bearing).
PremisecitationIn [9] an adaptive control scheme was proposed to adapt the global FL aggregation frequency.
“In [9] an adaptive control scheme was proposed to adapt the global FL aggregation frequency.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [9] source
Not checked in the free run.
May combine two claims.
InterpretationreasoningFor a fixed global FL model accuracy, an increase in computational cost reduces communication cost and vice versa.
“Therefore, for a fixed global FL model accuracy, an increase in computational cost reduces communication cost and vice versa.”
Quote in the paperSubstantial claim
The paper relies on this without support.
Federated learning can be a promising solution for enabling IoT-based smart applications.
“Federated learning can be a promising solution for enabling IoTbased smart applications.”
The paper relies on this without support.
UEs are generally reluctant to participate in FL due to limited computing and communication resources.
“However, UEs are generally reluctant to participate in FL due to limited computing and communication resources.”
The authors state this as an assumption.
The utility of the BS is modeled as a strictly concave function of key FL performance metrics such as the number of global iterations required to reach global accuracy for a given local relative accuracy.
“The utility of the BS is modeled as strictly concave function of key FL performance metrics such as the number of global iterations required to reach global accuracy for a given local relative accuracy.”
The authors state this as an assumption.
For the UEs utility, we choose a concave function of the local relative accuracy and the BS-offered reward.
“For the UEs utility, we choose a concave function of the local relative accuracy and the BS-offered reward.”
The paper relies on this without support.
The allocation of local device computational resources strongly depends on device energy consumption, local learning time, and local learning accuracy.
“The allocation of local device computational resources strongly depends on the device energy consumption, local learning time, and local learning accuracy.”
The paper relies on this without support.
Heterogeneous UEs have different computational and communication costs for training and, thus, they expect different rewards.
“However, heterogeneous UEs have different computational and communication costs for training and, thus, they expect different rewards.”
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?
FL can assist to mitigate communication expenses, such as data lag and power transmission, associated with transfer of raw data by avoiding the offloading of large data volumes to the server.
“Health Data Training at a Low Cost: FL can assist to mitigate communication expenses, such as data lag and power transmission, associated with transfer of raw data by avoiding the offloading of large data volumes to the server [97].”
From Federated learning-based AI approaches in smart healthcare: Concepts, taxonomies, challenges and open issues (2023) · cites this paper as [97]
FL-AI may assist decrease communication costs, like latency and transmit power, associated with raw data transfer by evading the dumping of large data to the central network because model gradients are often considerably less in size than their real datasets.
“FL-AI may assist decrease communication costs, like latency and transmit power, associated with raw data transfer by evading the dumping of large data to the central network because model gradients are often considerably less in size than their real datasets [97].”
From Federated learning-based AI approaches in smart healthcare: Concepts, taxonomies, challenges and open issues (2023) · cites this paper as [97]
By avoiding offloading huge data volumes to the server, FL reduces communication costs, such as latency and transmit power, consumed by raw data transmission, as model gradients generally have much smaller sizes compared to their actual datasets.
“By avoiding the offloading of huge data volumes to the server, FL can help reduce significantly communication costs, e.g., latency and transmit power, consumed by raw data transmission, as the model gradients generally have much smaller sizes compared to their actual datasets [32].”
From Federated learning for smart healthcare: A survey (2023) · cites this paper as [32]
Game theory and blockchain are two important tools for designing incentive mechanisms for federated learning.
“Game theory and blockchain are two important tools for designing incentive mechanisms”
From Federated learning for smart healthcare: A survey (2023) · cites this paper as [32]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Figure 3a identifies the wrong x-axis
MinorThe text calls relative accuracy the x-axis variable in Fig. 3a, but the figure labels the x-axis "Reward rate" and relative accuracy on the other axis.
“The accuracy improves when the relative accuracy value (x-axis) is smaller.”
“0.0 0.2 0.4 0.6 0.8 1.0 Reward rate 0.0 0.2 0.4 0.6 0.8 1.0 Relative accuracy Client 1 Client 2 Client 3 Client 4 Client 5 (a)”
Cites
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 · see citing works
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 · see citing works
Self-citationFederated 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 · see citing works
Adaptive 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 · see citing works
A Joint Learning and Communications Framework for Federated Learning over Wireless Networks
Chen, M., Yang, Z., Saad, W., Yin, C., Poor, H. V., & Cui, S. (2019) · arXiv · doi:10.48550/arxiv.1909.07972 · Cited by 1,547 · see citing works
Self-citationIn-Edge AI: Intelligentizing Mobile Edge Computing, Caching and Communication by Federated Learning
Wang, X., Han, Y., Wang, C., Zhao, Q., Chen, X., & Chen, M. (2019) · IEEE Network · doi:10.1109/mnet.2019.1800286 · Cited by 1,030 · see citing works
Federated Learning of Deep Networks using Model AveragingPreprint
McMahan, H. B., Moore, E., Ramage, D., & y Arcas, B. A. (2016) · arXiv (Cornell University) · Cited by 923 · see citing works
Chen, M., Mozaffari, M., Saad, W., Yin, C., Debbah, M., & Hong, C. S. (2017) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2017.2680898 · Cited by 742 · see citing works
Self-citationFair Resource Allocation in Federated Learning
Li, T., Sanjabi, M., Beirami, A., & Smith, V. (2019) · arXiv · doi:10.48550/arxiv.1905.10497 · Cited by 475 · see citing works
A Performance Evaluation of Federated Learning Algorithms
Nilsson, A., Smith, S., Ulm, G., Gustavsson, E., & Jirstrand, M. (2018) · Proceedings of the Second Workshop on Distributed Infrastructures for Deep Learning · doi:10.1145/3286490.3286559 · Cited by 328 · see citing works
A Crowdsourcing Framework for On-Device Federated Learning
Pandey, S. R., Tran, N. H., Bennis, M., Tun, Y. K., Manzoor, A., & Hong, C. S. (2020) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2020.2971981 · Cited by 318 · see citing works
Self-citationCommunication-Efficient Distributed Dual Coordinate AscentPreprint
Jaggi, M., Smith, V., Takáč, M., Terhorst, J., Krishnan, S., Hofmann, T., & Jordan, M. I. (2014) · arXiv (Cornell University) · doi:10.48550/arxiv.1409.1458 · Cited by 203 · see citing works
Artificial intelligence-based semantic internet of things in a user-centric smart city
Guo, K., Lu, Y., Gao, H., & Cao, R. (2018) · Sensors · doi:10.3390/s18051341 · Cited by 104 · see citing works
Wireless Network Intelligence at the Edge
Park, J., Samarakoon, S., Bennis, M., & Debbah, M. (2019) · Proceedings of the IEEE · doi:10.1109/jproc.2019.2941458 · Cited by 92 · see citing works
Software-Hardware Codesign for Efficient Neural Network Acceleration
Guo, K., Han, S., Yao, S., Wang, Y., Xie, Y., & Yang, H. (2017) · IEEE Micro · doi:10.1109/mm.2017.39 · Cited by 78 · cited 1× · see citing works
“Therefore, there is a need to jointly consider both hardware design space and neural architecture search space for a more flexible design of the end device for FL [12].”
BackgroundWireless Edge Computing With Latency and Reliability Guarantees
Elbamby, M. S., Perfecto, C., Liu, C., Park, J., Samarakoon, S., Chen, X., & Bennis, M. (2019) · Proceedings of the IEEE · doi:10.1109/jproc.2019.2917084 · Cited by 17 · see citing works
Flagged first, then by citations
Research works printed but not linked: 3
- smith, 2017, Federated Multi-Task Learning, Proc Adv Neural Info Process Syst, 4424
- han, 2012, Game Theory in Wireless and Communication Networks Theory Models and Applications
- dinh, 2019, Federated learning over wireless networks Convergence analysis and resource allocation
Cited by
Sorted by citations
Federated 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 · see citing works
A review of applications in federated learning
Li, L., Fan, Y., Tse, M., & Lin, K.-Y. (2020) · Computers & Industrial Engineering · doi:10.1016/j.cie.2020.106854 · Cited by 1,559 · see citing works
A survey on security and privacy of federated learning
Mothukuri, V., Parizi, R. M., Pouriyeh, S., Huang, Y., Dehghantanha, A., & Srivastava, G. (2021) · Future Generation Computer Systems · doi:10.1016/j.future.2020.10.007 · Cited by 1,480 · see citing works
Federated learning for smart healthcare: A survey
Nguyen, D. C., Pham, Q.-V., Pathirana, P. N., Ding, M., Seneviratne, A., Lin, Z., Dobre, O., & Hwang, W.-J. (2023) · ACM Computing Surveys · doi:10.1145/3501296 · Cited by 827 · see citing works
A Survey on Federated Learning for Resource-Constrained IoT Devices
Imteaj, A., Thakker, U., Wang, S., Li, J., & Amini, M. H. (2022) · IEEE Internet of Things Journal · doi:10.1109/jiot.2021.3095077 · Cited by 774 · see citing works
Federated learning for 6G communications: Challenges, methods, and future directions
Liu, Y., Yuan, X., Xiong, Z., Kang, J., Wang, X., & Niyato, D. (2020) · China Communications · doi:10.23919/jcc.2020.09.009 · Cited by 464 · see citing works
Rahman, A., Hossain, M. S., Muhammad, G., Kundu, D., Debnath, T., Rahman, M., Khan, M. S. I., Tiwari, P., & Band, S. S. (2023) · Cluster Computing · doi:10.1007/s10586-022-03658-4 · Cited by 399 · see citing works
Federated Learning for Privacy Preservation in Smart Healthcare Systems: A Comprehensive Survey
Ali, M., Naeem, F., Tariq, M., & Kaddoum, G. (2023) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2022.3181823 · Cited by 366 · see citing works
Blockchain-empowered Federated Learning: Challenges, Solutions, and Future Directions
Zhu, J., Cao, J., Saxena, D., Jiang, S., & Ferradi, H. (2023) · ACM Computing Surveys · doi:10.1145/3570953 · Cited by 309 · see citing works
Blockchain-enabled Federated Learning: A Survey
Qu, Y., Uddin, M. P., Gan, C., Xiang, Y., Gao, L., & Yearwood, J. (2023) · ACM Computing Surveys · doi:10.1145/3524104 · Cited by 286 · see citing works
Federated learning for secure IoMT-applications in smart healthcare systems: A comprehensive review
Rani, S., Kataria, A., Kumar, S., & Tiwari, P. (2023) · Knowledge-Based Systems · doi:10.1016/j.knosys.2023.110658 · Cited by 228 · see citing works
Challenges, Applications and Design Aspects of Federated Learning: A Survey
Rahman, K. M. J., Ahmed, F., Akhter, N., Hasan, M., Amin, R., Aziz, K. E., Islam, A. K. M. M., Mukta, M. S. H., & Islam, A. K. M. N. (2021) · IEEE Access · doi:10.1109/access.2021.3111118 · Cited by 147 · see citing works
Federated learning for healthcare applications
Chaddad, A., Wu, Y., & Desrosiers, C. (2024) · IEEE Internet of Things Journal · doi:10.1109/jiot.2023.3325822 · Cited by 136 · see citing works
Challenges and future directions of secure federated learning: a survey
Zhang, K., Song, X., Zhang, C., & Yu, S. (2022) · Frontiers of Computer Science · doi:10.1007/s11704-021-0598-z · Cited by 127 · see citing works
Low-Latency Federated Learning Over Wireless Channels With Differential Privacy
Wei, K., Li, J., Ma, C., Ding, M., Chen, C., Jin, S., Han, Z., & Poor, H. V. (2022) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2021.3126052 · Cited by 109 · see citing works
Self-citationPrivacy-Enhancing Technologies in Federated Learning for the Internet of Healthcare Things: A Survey
Mosaiyebzadeh, F., Pouriyeh, S., Parizi, R. M., Sheng, Q. Z., Han, M., Zhao, L., Sannino, G., Ranieri, C. M., Ueyama, J., & Batista, D. M. (2023) · Electronics · doi:10.3390/electronics12122703 · Cited by 51 · see citing works
Liu, Y., Liu, P., Jing, W., & Song, H. H. (2023) · IEEE Internet of Things Journal · doi:10.1109/jiot.2023.3295763 · Cited by 36 · see citing works
Blockchain-Inspired Collaborative Cyber-Attacks Detection for Securing Metaverse
Zainudin, A., Putra, M. A. P., Alief, R. N., Akter, R., Kim, D.-S., & Lee, J.-M. (2024) · IEEE Internet of Things Journal · doi:10.1109/jiot.2024.3364247 · Cited by 35 · see citing works
Privacy Preservation for the IoMT Using Federated Learning and Blockchain Technologies
Alalawi, S., Alalawi, M., & Alrae, R. (2024) · Lecture Notes in Networks and Systems · doi:10.1007/978-3-031-65522-7_62 · Cited by 6 · see citing works
- Authors
- Khan, L. U., Pandey, S. R., Tran, N. H., Saad, W., Han, Z., Nguyen, M. N. H., & Hong, C. S.
- Affiliations
- Kyung Hee University (KR)
- The University of Sydney (AU)
- Virginia Tech (US)
- University of Houston (US)
- ORCID
- Venue
- IEEE Communications Magazine
- Published
- 2020-10
- Volume
- 58
- Issue
- 10
- Pages
- 88-93
- ISSN
- 0163-6804 (ISSN-L)
- 1558-1896
- OpenAlex
- W2985145906
- Cited by
- 506
- References
- 16
- Citations by year
- 2026: 37 · 2025: 57 · 2024: 87 · 2023: 106 · 2022: 102 · 2021: 92 · 2020: 25
- FWCI
- 31.94
- Type
- journal article
- Language
- en
- Open access
- green
- arxiv.org
- Licence
- Locations
- IEEE Communications Magazine (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- VBN Forskningsportal (Aalborg Universitet) (submittedVersion)
- arXiv (Cornell University)
- 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 › Computer Science › Information Systems › Internet of Things and AI
- Keywords
- federated learning, edge computing, Internet of Things, resource optimization, incentive mechanisms, Stackelberg game, resource allocation
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.
Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-09.
Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-09.
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
