Wireless Network Intelligence at the Edge
Open accessThe DOI matches its Crossref record; 205 of the 224 references it cites were found in a registry or an index, and 6 are not checked yet.
Abstract
Fueled by the availability of more data and computing power, recent breakthroughs in cloud-based machine learning (ML) have transformed every aspect of our lives from face recognition and medical diagnosis to natural language processing. However, classical ML exerts severe demands in terms of energy, memory, and computing resources, limiting their adoption for resource-constrained edge devices. The new breed of intelligent devices and high-stake applications (drones, augmented/virtual reality, autonomous systems, and so on) requires a novel paradigm change calling for distributed, low-latency and reliable ML at the wireless network edge (referred to as edge ML). In edge ML, training data are unevenly distributed over a large number of edge nodes, which have access to a tiny fraction of the data. Moreover, training and inference are carried out collectively over wireless links, where edge devices communicate and exchange their learned models (not their private data). In a first of its kind, this article explores the key building blocks of edge ML, different neural network architectural splits and their inherent tradeoffs, as well as theoretical and technical enablers stemming from a wide range of mathematical disciplines. Finally, several case studies pertaining to various high-stake applications are presented to demonstrate the effectiveness of edge ML in unlocking the full potential of 5G and beyond.
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?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Examples of adversarial attacks include data poisoning [69, 319], model update poisoning [44, 67], and model evasion attacks [441, 69, 211].”
From Advances and Open Problems in Federated Learning (2021) · cites this paper as [211]
Experimental — accuracy not yet measured
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
Hochreiter, S., & Schmidhuber, J. (1997) · Neural Computation · doi:10.1162/neco.1997.9.8.1735 · Cited by 102,318 · cited 1×
“To cope with this, the vanilla RNN’s hidden layer can be replaced with a long short-term memory (LSTM) unit [74].”
BackgroundElements of Information Theory
Cover, T. M., & Thomas, J. A. (2001) · Wiley eBooks · doi:10.1002/0471200611 · Cited by 36,808
(2014) · Cited by 33,419
Learning Phrase Representations using RNN Encoder–Decoder for Statistical Machine Translation
Cho, K., van Merrienboer, B., Gulcehre, C., Bahdanau, D., Bougares, F., Schwenk, H., & Bengio, Y. (2014) · Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.3115/v1/d14-1179 · Cited by 25,545 · cited 5×
“GRUs, a gating mechanism in RNNs, are commonly used for modeling speech signals and musics.”
BackgroundInfluentialMultilayer feedforward networks are universal approximators
Hornik, K., Stinchcombe, M., & White, H. (1989) · Neural Networks · doi:10.1016/0893-6080(89)90020-8 · Cited by 21,728 · cited 29×
“Here, n is the sample complexity to satisfy a target generalization error with a target reliability 1 − ε. PAC Framework Applications: (a) PAC-VC Bound: If the hypothesis space H is infinite, e.g., in deep NNs or non-parametric models such as GP, one can derive the GE using the VapnikChervonenkis (VC) dimension VC(H).”
BackgroundInfluentialCommunication-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 4×
“FL has been shown to improve communication efficiency by MSI quantization [30], adjusting the MSI update period [28], and optimizing devices’ scheduling policy [31].”
BackgroundMethodInfluentialDistilling the Knowledge in a Neural Network
Hinton, G., Vinyals, O., & Dean, J. (2015) · arXiv · doi:10.48550/arxiv.1503.02531 · Cited by 14,204
Approximation by superpositions of a sigmoidal function
Cybenko, G. (1989) · Mathematics of Control, Signals, and Systems · doi:10.1007/bf02551274 · Cited by 13,883 · cited 1×
“The proof is provided for shallow NNs [83], [84] and recently for deep NNs [85].”
BackgroundFoundations of statistical natural language processing
Manning, C. D., & Schütze, H. (1999) · Cited by 9,816
Playing Atari with Deep Reinforcement Learning
Mnih, V., Kavukcuoglu, K., Silver, D., Graves, A., Antonoglou, I., Wierstra, D., & Riedmiller, M. (2013) · arXiv · doi:10.48550/arxiv.1312.5602 · Cited by 9,471 · cited 1×
“To illustrate its effectiveness, consider the Atari gaming environment [190] whose output action dimension is only {up, down, left, right}, whereas the input state dimension is the entire pixels per frame.”
Research works printed but not linked: 30
- 2018, Apple iPhone Xs Max
- 2018, Snapdragon 845 Mobile Platform
- 2018, Comparison Chart
- oulu, 2018, 6G Flagship
- 2018, Cloud TPU
- panda, 2018, RDMA-Based Networking Technologies and Middleware for Next-Generation Clusters and Data Centers
- 2018, Architecture Specification
- 2016, Feasibility Study on New Services and Markets Technology Enablers Release-14
- 2017, Study on New Radio Access Technology Physical Layer Aspects Release-14
- 2018, Study on Scenarios and Requirements for Next Generation Access Technologies Release 14
- iwabuchi, 2018, Field experiments on 5G ultra-reliable low-latency communication (URLLC), NTT DoCoMo Tech J, 20, 14
- 2017, Minimum Requirements Related to Technical Performance for IMT-2020 Radio Interface(s)
- 2018, Google Pixel 3 XL
- 10.1109/glocom.2016.7842367 · doi:10.1109/glocom.2016.7842367
- 2016, Recommendations for NGMN KPIs and Requirements for 5G
- briman, 2018, Another Look at Alphago vs Lee Sedol The Power Angle
- goodfellow, 2016, Deep Learning
- 2018, Service Requirements for the 5G System Stage 1 Release-14
- chaudhari, 2017, Entropy-SGD: Biasing gradient descent into wide valleys, Proc ICLR, 1
- vepakomma, 2019, Reducing leakage in distributed deep learning for sensitive health data, Proc ICLR, 1
- nakamoto, 2018, Bitcoin A Peer-to-Peer Electronic Cash System
- 2017, Coverage Analysis of LTE-M Category-M1
- metz, 2018, Machine Learning Invades the Real World on Internet Balloons
- weisstein, 2018, Fractional Chromatic Number
- 2015, IMT vision—Framework and overall objectives of the future development of IMT for 2020 and beyond
- 2017, Study on New Radio Access Technology Radio Access Architecture and Interfaces Release-14
- elgabli, 2019, Communication-efficient decentralized machine learning framework for 5G and beyond, Proc Globecom, 1
- 2018, Intel Skylake X
- 2018, Intel SSD DC P3700 Review
- 2017, Augmented and Virtual Reality The First Wave of 5G Killer Apps
Cited by
Items 301–400 of 702
Heterogeneous Coded Computation across Heterogeneous Workers
Sun, Y., Zhao, J., Zhou, S., & Gündüz, D. (2019) · Globecom · doi:10.1109/globecom38437.2019.9014006 · Cited by 21
“The straggling parameter u m,n is randomly selected within [1 , 5] ms − 1 , the shift parameter is set as a m,n = 1 u m,n ms , and L m = 10 5 , ∀ m [5].”
BackgroundHuang, S., Zhang, H., Wang, X., Chen, M., Li, J., & Leung, V. C. M. (2022) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2022.3226448 · Cited by 20
“Compared to reacting to users’ requests passively, such network intelligence technology enables operators and mobile service providers to proactively improve their resource provision or cache strategies to achieve higher network throughput and lower service latency [4].”
BackgroundCommunication Efficient Framework for Decentralized Machine Learning
Elgabli, A., Park, J., Bedi, A. S., Bennis, M., & Aggarwal, V. (2020) · Conference on Information Sciences and Systems · doi:10.1109/ciss48834.2020.1570627384 · Cited by 20
“For a large scale network, the communication cost often becomes dominant compared to the computation time, calling for communication efficient distributed optimization [16], [17], [18], [19], [20], [21].”
BackgroundSelf-citationCost-efficient Distributed optimization In Machine Learning Over Wireless Networks
Mahmoudi, A., Shokri‐Ghadikolaei, H., & Fischione, C. (2020) · International Conference on Communications · doi:10.1109/icc40277.2020.9149216 · Cited by 20
“Use cases include machine leaning over Internet-of-Things, edge computing, vehicular networks, mobile communications, or general networks with background traffic [2], [3].”
MethodYoon, J., Byeon, Y., Kim, J., & Lee, H. (2021) · IEEE Internet of Things Journal · doi:10.1109/jiot.2021.3131407 · Cited by 19
“network of resource-constrained edge devices [18], [19].”
Ahn, J.-H., Bennis, M., & Kang, J. (2022) · IEEE Transactions on Green Communications and Networking · doi:10.1109/tgcn.2022.3186538 · Cited by 19
“Federated Learning (FL) was recently proposed as a privacy-preserving distributed ML scheme based on first-order distributed optimization updates and periodic exchanges of model parameter information between devices (or clients) and a parameter server (PS) [1]–[5].”
BackgroundSelf-citationUnleashing Edgeless Federated Learning With Analog Transmissions
Yang, H. H., Chen, Z., & Quek, T. Q. S. (2024) · IEEE Transactions on Signal Processing · doi:10.1109/tsp.2024.3352405 · Cited by 19
Han, P., Liu, Y., Zhang, X., & Guo, L. (2022) · IEEE Transactions on Vehicular Technology · doi:10.1109/tvt.2022.3194543 · Cited by 19
Multi-User Goal-Oriented Communications With Energy-Efficient Edge Resource Management
Binucci, F., Banelli, P., Lorenzo, P. D., & Barbarossa, S. (2023) · IEEE Transactions on Green Communications and Networking · doi:10.1109/tgcn.2023.3275199 · Cited by 19
“In this view, the allocation of system resources in order to reach prescribed target performance in terms of latency, accuracy, and energy consumption has been already considered in [3]–[6].”
BackgroundCommunication-Efficient and Personalized Federated Lottery Ticket Learning
Seo, S., Ko, S., Park, J., Kim, S., & Bennis, M. (2021) · IEEE International Workshop on Signal Processing Advances in Wireless Communications (SPAWC) · doi:10.1109/spawc51858.2021.9593126 · Cited by 18
Self-citationPower allocation strategies for 6G communication in VL-NOMA systems: an overview
Ngene, C. E., Thakur, P., & Singh, G. K. (2023) · Smart Science · doi:10.1080/23080477.2023.2225944 · Cited by 18
“In VL-NOMA, ML will be deployed because it improves the performance of delay-sensitive applications enablers; the researcher has to exploit ML algorithms at the edge of the network [61].”
MethodMix2SFL: Two-Way Mixup for Scalable, Accurate, and Communication-Efficient Split Federated Learning
Oh, S., Nam, H., Park, J., Vepakomma, P., Raskar, R., Bennis, M., & Kim, S. (2023) · IEEE Transactions on Big Data · doi:10.1109/tbdata.2023.3328424 · Cited by 18
“U TILIZING large amounts of data through large-scale parallel computing power is instrumental in high-quality deep learning [1], [2], [3], [4].”
BackgroundSelf-citationAvailability-Aware Revenue-Effective Application Deployment in Multi-Access Edge Computing
Zhao, L., Xiao, F., Li, B., Zhou, J., Xu, X., & Yang, Y. J. (2024) · IEEE Transactions on Parallel and Distributed Systems · doi:10.1109/tpds.2024.3399840 · Cited by 18
“Moreover, compared to wired transmissions, wireless communications may fail occasionally [15], [16], [17], e.g., the unpredictable environmental effects can make wireless links unreliable [18].”
BackgroundUnderstanding UAV-Based WPCN-Aided Capabilities for Offshore Monitoring Applications
Carrillo, D., Mikhaylov, K., Nardelli, P. J., Andreev, S., & da Costa, D. B. (2021) · IEEE Wireless Communications · doi:10.1109/mwc.001.2000218 · Cited by 18
“These expenditures may be particularly significant if data processing is handled by a remote cloud, which can be complemented by relying on the edge computing capabilities to improve the processing efficiency and reduce the costs [13].”
BackgroundEDISON: An Edge-Native Method and Architecture for Distributed Interpolation
Lovén, L., Lähderanta, T., Ruha, L., Peltonen, E., Launonen, I. K., Sillanpää, M. J., Riekki, J. P., & Pirttikangas, S. (2021) · Sensors · doi:10.3390/s21072279 · Cited by 18
“The current, heavily cloud-based solutions require data aggregation and processing in a remote computational environment, imposing several challenges such as high networking load and latency, high transmission costs, and loss of privacy [16,21,22].”
BackgroundJang, H., Skatchkovsky, N., & Simeone, O. (2021) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr48806.2021.9413206 · Cited by 18
“of ANN-based solutions in terms of energy consumption [3], particularly for edge intelligence applications [4], there is a renewed interest in exploring novel computational paradigms that can benefit from time encoding and temporally sparse, event-driven, processing [5].”
BackgroundDynamic Resource Optimization for Adaptive Federated Learning at the Wireless Network Edge
Lorenzo, P. D., Battiloro, C., Merluzzi, M., & Barbarossa, S. (2021) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp39728.2021.9414832 · Cited by 18
Systematic Review of Virtual Reality Solutions Employing Artificial Intelligence Methods
de Oliveira, T. R., da Silva, M. M., Spinasse, R. A. N., Ludke, G. G., Gaudio, M. R. S., Gomes, G. I. R., Cotini, L. G., Vargens, D. D. S., Schimidt, M. Q., Andreão, R. V., et al. (2021) · Symposium on Virtual and Augmented Reality (SVR) · doi:10.1145/3488162.3488209 · Cited by 18
Optimizing Pipelined Computation and Communication for Latency-Constrained Edge Learning
Skatchkovsky, N., & Simeone, O. (2019) · IEEE Communications Letters · doi:10.1109/lcomm.2019.2922658 · Cited by 18
A 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
“…the reliability or the FL and RL Optimization of the edge computing (mobile) and for requirements of caching [171] FL FL integration and challenges [172] Edge-ML Illustrations of the architecture of Edge-ML [173] FL Integration and characteristics of FL [174] FL Use cases/ Application and Services…”
LinguaLinked: Distributed Large Language Model Inference on Mobile Devices
Zhao, J., Song, Y., Harris, I. G., & Jyothi, S. A. (2024) · Proceedings of the conference - Association for Computational Linguistics. Meeting · doi:10.18653/v1/2024.acl-demos.16 · Cited by 17
Wu, H., Wang, P., & Narayana, C. V. A. (2025) · IEEE Transactions on Network Science and Engineering · doi:10.1109/tnse.2025.3577910 · Cited by 17
“Without accessing user-sensitive data, user privacy is protected, and the communication cost is reduced, making FL one of the most promising technologies for future network intelligence [3].”
Robustness and Diversity Seeking Data-Free Knowledge Distillation
Han, P., Park, J., Wang, S., & Liu, Y. (2021) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp39728.2021.9414674 · Cited by 17
“Machine intelligence has been increasingly trickling down from powerful cloud servers to edge devices such as phones, cars, and Internet of things (IoT) [1, 2, 3].”
BackgroundSelf-citationEdge Intelligence in 6G Systems
Chaccour, C., & Saad, W. (2021) · Computer communications and networks · doi:10.1007/978-3-030-72777-2_12 · Cited by 17
Completion Time Minimization of Fog-RAN-Assisted Federated Learning With Rate-Splitting Transmission
Park, S.-H., & Lee, H. (2022) · IEEE Transactions on Vehicular Technology · doi:10.1109/tvt.2022.3180747 · Cited by 17
“To avoid long latency required for exchange of local and global model updates between IDs and CS, edge learning has been actively studied, which enjoys low latency by bringing the decision-making tasks closer to IDs [7].”
BackgroundDynamic Resource Allocation for Multi-User Goal-oriented Communications at the Wireless Edge
Binucci, F., Banelli, P., Lorenzo, P. D., & Barbarossa, S. (2022) · European Signal Processing Conference (EUSIPCO) · doi:10.23919/eusipco55093.2022.9909863 · Cited by 17
“To this aim, edge machine learning (EML) is becoming a hot research topic [2]-[5], enabling mobile devices (MDs) to opportunistically offload learning tasks to ESs.”
BackgroundToward Massive Distribution of Intelligence for 6G Network Management Using Double Deep Q-Networks
Majumdar, S., Schwarzmann, S., Trivisonno, R., & Carle, G. (2023) · IEEE Transactions on Network and Service Management · doi:10.1109/tnsm.2023.3333875 · Cited by 17
“It leverages the enormous data available and derives optimal decisions in real-time [9].”
BackgroundCommunication Efficient Federated Learning
Jin, Y., Zhu, H., Xu, J., & Chen, Y. (2022) · Machine learning · doi:10.1007/978-981-19-7083-2_2 · Cited by 17
RingFed: Reducing Communication Costs in Federated Learning on Non-IID DataPreprint
Yang, G., Ke, M., Song, C., Yang, Z., & Gong, T. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2107.08873 · Cited by 17
“Therefore, the single central server is separated into several edge servers to process data close to the clients to reduce time delays [6].”
A Survey on Federated Learning PoisoningAttacks and DefensesPreprint
Liang, J., & Rong, W. (2022) · Research Square · doi:10.21203/rs.3.rs-1900743/v1 · Cited by 17
Kanagarla, K. P. B. (2024) · SSRN Electronic Journal · doi:10.2139/ssrn.5012466 · Cited by 17
FedCau: A Proactive Stop Policy for Communication and Computation Efficient Federated Learning
Mahmoudi, A., Ghadikolaei, H. S., Barros Da Silva, J. M., & Fischione, C. (2024) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2024.3378351 · Cited by 16
“Moreover, they neglect other crucial costs associated with solving federated learning (FL) problems, such as latency [3] and energy consumption [20].”
BackgroundMethodSliding Differential Evolution Scheduling for Federated Learning in Bandwidth-Limited Networks
Luo, Y., Xu, J., Xu, W., & Wang, K. (2020) · IEEE Communications Letters · doi:10.1109/lcomm.2020.3032517 · Cited by 16
“, access points (APs), edge networks can be established and are able to conduct complex task via intelligent scheduling and processing [1].”
BackgroundSoorki, M. N., Saad, W., Bennis, M., & Hong, C. S. (2021) · IEEE Transactions on Communications · doi:10.1109/tcomm.2021.3106686 · Cited by 16
Self-citationLiu, J., Xu, Z., & Wen, Z. (2023) · IEEE Transactions on Vehicular Technology · doi:10.1109/tvt.2023.3265479 · Cited by 16
EdgeMove: Pipelining Device-Edge Model Training for Mobile Intelligence
Dong, Z., He, Q., Chen, F., Jin, H., Gu, T., & Yang, Y. J. (2023) · International World Wide Web Conference (WWW) · doi:10.1145/3543507.3583540 · Cited by 16
“However, it is difficult and often impossible to obtain a high model accuracy with these techniques [65].”
BackgroundThe first verification test of space-ground collaborative intelligence via cloud-native satellites
Shangguang, W., Zhang, Q., Xing, R., Qi, F., & Xu, M. (2024) · China Communications · doi:10.23919/jcc.fa.2022-0422.202404 · Cited by 16
“Such large data transmission to the ground station may take up most of the time and induce huge communication overhead under the bent-pipe architecture [17].”
BackgroundAn Efficient Distributed Deep Learning Framework for Fog-Based IoT Systems
Chang, Y., Huang, X., Shao, Z., & Yang, Y. (2019) · Globecom · doi:10.1109/globecom38437.2019.9014056 · Cited by 16
“However, IoT devices and fog nodes are usually resource-constrained, which calls for a distributed design of DNN deployment across IoT devices and fog nodes [3].”
BackgroundFedCo: A Federated Learning Controller for Content Management in Multi-party Edge Systems
Balasubramanian, V., Aloqaily, M., & Martin Reisslein, B. (2021) · International Conference on Computer Communications and Networks (ICCCN) · doi:10.1109/icccn52240.2021.9522153 · Cited by 16
Noise Is Useful: Exploiting Data Diversity for Edge Intelligence
Zeng, Z., Liu, Y., Tang, W., & Chen, F. (2021) · IEEE Wireless Communications Letters · doi:10.1109/lwc.2021.3051688 · Cited by 15
“urges efficient wireless data transmission solutions for edge intelligence [4]–[7].”
BackgroundOptimized Power Control for Over-the-Air Federated Edge Learning
Cao, X., Zhu, G., Xu, J., & Cui, S. R. (2021) · International Conference on Communications · doi:10.1109/icc42927.2021.9500664 · Cited by 15
“INTRODUCTION In the pursuit of ubiquitous intelligence envisioned in the future 6G networks, recent years have witnessed the spreading of artificial intelligence (AI) algorithms from the cloud to the network edge, resulting in an active area called edge intelligence [1], [2].”
BackgroundPredictive Closed-Loop Remote Control Over Wireless Two-Way Split Koopman Autoencoder
Girgis, A. M., Seo, H., Park, J., Bennis, M., & Choi, J. (2022) · IEEE Internet of Things Journal · doi:10.1109/jiot.2022.3206415 · Cited by 15
“In particular, we consider an AE split into its encoder and decoder parts, whereby the smashed data or latent representations propagate from the encoder to the decoder while the prediction errors or gradients propagate back in the reverse direction [20], [21].”
MethodSelf-citationDistributed Signal Strength Prediction using Satellite Map empowered by Deep Vision Transformer
Yu, H., Hou, Z., Gu, Y., Cheng, P., Ouyang, W., Li, Y., & Vucetic, B. S. (2021) · IEEE Globecom Workshops (GC Wkshps) · doi:10.1109/gcwkshps52748.2021.9682021 · Cited by 15
“1 How to select participated users [9] is an interesting problem deserving further study.”
BackgroundChallenges of AI in Wireless Networks for IoTPreprint
Ahmad, I., Shahabuddin, S., Kumar, T., Harjula, E., Meisel, M., Juntti, M. J., Sauter, T., & Ylianttila, M. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2007.04705 · Cited by 15
“If resources are available in high-end servers in centralized cloud systems, latency critical applications will be challenged by the communication latency, besides other factors [7].”
BackgroundFederated Reinforcement Distillation with Proxy Experience MemoryPreprint
Cha, H., Park, J., Kim, H., Kim, S., & Bennis, M. (2020) · doi:10.36227/techrxiv.12645497 · Cited by 15
Self-citation6G and V2X Communications: Applications, Features, and Challenges
Ouamna, H., Madini, Z., & Zouine, Y. (2022) · 2022 8th International Conference on Optimization and Applications (ICOA) · doi:10.1109/icoa55659.2022.9934407 · Cited by 14
“Federated learning, which has emerged in recent years to address privacy and communication overload issues associated with ML model formation, has sparked much research interest in improved wireless networks [39].”
BackgroundHarnessing wireless channels for scalable and privacy-preserving federated learning
Elgabli, A., Park, J., Issaid, C. B., & Bennis, M. (2021) · IEEE Transactions on Communications · doi:10.1109/tcomm.2021.3078783 · Cited by 14
Self-citation6G Goal-Oriented Communications: How to Coexist with Legacy Systems?
(2024) · doi:10.3390/telecom5010005 · Cited by 14
CaBaFL: Asynchronous Federated Learning via Hierarchical Cache and Feature Balance
Xia, Z., Hu, M., Yan, D., Xie, X., Li, T., Li, A., Zhou, J., & Chen, M. (2024) · IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems · doi:10.1109/tcad.2024.3446881 · Cited by 14
Mixture-of-Experts for Distributed Edge Computing with Channel-Aware Gating Function
Song, Q., Jing, S., Zhang, S., Zhang, S., & Huang, C. Z. (2025) · IEEE International Conference on Communications workshops · doi:10.1109/iccworkshops67674.2025.11162260 · Cited by 14
“Given its inherently distributed design, MoE is a promising approach for allocating computational resources across multiple devices in distributed wireless communication systems [5].”
FlexSpec: Frozen Drafts Meet Evolving Targets in Edge-Cloud Collaborative LLM Speculative Decoding
Li, Y., Kong, R., Lyu, Z., Li, Q., Chen, X., Cai, H., Yan, L., Wang, S., Zhao, J., Zhu, G., et al. (2026) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2026.3683574 · Cited by 14
“To tackle such issues, Park et al. [30] and Zhu et al. [31] laid the groundwork for wireless edge intelligence by optimizing the trade-off between computation offloading and communication channel quality.”
6G CloudNet: Towards a Distributed, Autonomous, and Federated AI-Enabled Cloud and Edge Computing
Alimi, I. A., Patel, R. K., Zaouga, A., Muga, N. J., Pinto, A. N., Teixeira, A., & Monteiro, P. P. (2021) · Computer communications and networks · doi:10.1007/978-3-030-72777-2_13 · Cited by 14
CFLIT: 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
“…services, such as autonomous driving and intelligent personal assistants, triggers the emergence of a new paradigm called edge intelligence [1], [2], [3], which migrates the training and inference processes of AI models from central cloud servers to the edge of wireless networks, i.e., edge…”
BackgroundPipeEdge: A Trusted Pipelining Collaborative Edge Training based on Blockchain
Yuan, L., He, Q., Chen, F., Dou, R., Jin, H., & Yang, Y. J. (2023) · International World Wide Web Conference (WWW) · doi:10.1145/3543507.3583413 · Cited by 14
“However, it is difficult and often impossible to obtain a high model accuracy with these techniques [43].”
BackgroundNetworking-Aware IoT Application Development
Bröring, A., Seeger, J., Papoutsakis, M., Fysarakis, K., & Caracalli, A. (2020) · Sensors · doi:10.3390/s20030897 · Cited by 14
Neural-Network-Based Root Mean Delay Spread Model for Ubiquitous Indoor Internet-of-Things Scenarios
Yu, Y., Lu, W., Liu, Y., & Zhu, H. (2020) · IEEE Internet of Things Journal · doi:10.1109/jiot.2020.2979766 · Cited by 14
“Actually, an alternative solution to satisfy the communication requirements in IoTs scenarios is to offload part of the communication and computing tasks to the edge of the networks [23], [24].”
BackgroundReal Time Arrhythmia Monitoring and Classification Based on Edge Computing and DNN
Liu, M., Shao, N., Zheng, C., & Wang, J. (2021) · Wireless Communications and Mobile Computing · doi:10.1155/2021/5563338 · Cited by 14
“6G wireless communication networks are believed that it should be ubiquitous, human-centric, full band, strongly secure, and intelligent [2, 3], which offers distributed, low latency, and reliable machine learning at the wireless network edge [4, 5].”
BackgroundDistributed Multi-Agent Learning for Service Function Chain Partial Offloading at the Edge
Khoramnejad, F., Joda, R., & Erol‐Kantarci, M. (2021) · IEEE International Conference on Communications workshops · doi:10.1109/iccworkshops50388.2021.9473554 · Cited by 14
“Based on the success of machine learning enabled solutions in 5G [10], using edge learning for such complex systems emerges as a promising approach [11], [12].”
BackgroundLearning Centric Power Allocation for Edge Intelligence
Wang, S., Wang, R., Hao, Q., Wu, Y., & Poor, H. V. (2020) · International Conference on Communications · doi:10.1109/icc40277.2020.9148872 · Cited by 14
CADA: Communication-Adaptive Distributed AdamPreprint
Chen, T., Guo, Z., Sun, Y., & Yin, W. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2012.15469 · Cited by 14
“[34] Jihong Park, Sumudu Samarakoon, Mehdi Bennis, and Mérouane Debbah.”
BackgroundMembership Inference Attacks via Adversarial ExamplesPreprint
Jalalzai, H., Kadoche, E., Leluc, R., & Plassier, V. (2022) · HAL · doi:10.48550/arxiv.2207.13572 · Cited by 14
“Sharing data and model in a private way is a major barrier to adoption of emerging AI technologies (Park et al., 2019).”
BackgroundRevolutionizing Wireless Networks with Federated Learning: A Comprehensive Review
Mahdimahalleh, S. E. (2024) · European Journal of Electrical Engineering and Computer Science · doi:10.24018/ejece.2024.8.6.671 · Cited by 13
“To address these challenges, a promising approach is to bring intelligence to the network edge by enabling wireless devices to implement distributed and collaborative learning algorithms [1, 2].”
Capacitated spatial clustering with multiple constraints and attributes
Lähderanta, T., Lovén, L., Ruha, L., Leppänen, T., Launonen, I. K., Riekki, J. P., & Sillanpää, M. J. (2023) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2023.107182 · Cited by 13
al., N. E. (2022) · International Journal of ADVANCED AND APPLIED SCIENCES · doi:10.21833/ijaas.2022.07.013 · Cited by 13
Koda, Y., Okura, R., & Harada, H. (2024) · IEEE Internet of Things Journal · doi:10.1109/jiot.2024.3399249 · Cited by 13
“Therein, the communicating nodes must immediately form an ad hoc WPAN and transmit a large amount of streaming data, locally exchange a large amount of sensory data, or transmit even machine-learning model parameters for control [21].”
BackgroundOptimized Amplify-and-Forward Relaying for Hierarchical Over-the-Air Computation
Wang, F., & Xu, J. (2020) · IEEE Globecom Workshops (GC Wkshps) · doi:10.1109/gcwkshps50303.2020.9367526 · Cited by 13
“INTRODUCTION The proliferation of smart wireless devices (WDs) has recently enabled various new applications such as wireless sensor networks, internet-of-things (IoT), and edge machine learning [1].”
BackgroundAge-Optimal Power Allocation in Industrial IoT: A Risk-Sensitive Federated Learning Approach
Hsu, Y.-L., Liu, C., Samarakoon, S., Wei, H., & Bennis, M. (2021) · IEEE International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC) · doi:10.1109/pimrc50174.2021.9569536 · Cited by 13
Self-citationAlimi, I. A., Patel, R. K., Zaouga, A., Muga, N. J., Xin, Q., Pinto, A. N., & Monteiro, P. P. (2021) · IntechOpen eBooks · doi:10.5772/intechopen.98315 · Cited by 13
“In view of this, federated learning can be employed to further enhance privacy [11, 40, 43].”
BackgroundStatistical Learning-Based Grant-Free Access for Delay-Sensitive Internet of Things Applications
Raza, M. A., Abolhasan, M., Lipman, J., Shariati, N., Ni, W., & Jamalipour, A. (2022) · IEEE Transactions on Vehicular Technology · doi:10.1109/tvt.2022.3155719 · Cited by 13
“…in heterogeneous networks should have the capability to adapt to network dynamics without relying much on the BS, which can be accomplished by equipping the IoT devices with reasonable computation resources, allowing them to learn the network parameters and share the knowledge withtheBS[24].”
BackgroundOnline Model Updating with Analog Aggregation in Wireless Edge Learning
Wang, J., Dong, M., Liang, B., Boudreau, G., & Abou-Zeid, H. (2022) · IEEE Conference on Computer Communications (INFOCOM) · doi:10.1109/infocom48880.2022.9796860 · Cited by 13
Joint Online Optimization of Model Training and Analog Aggregation for Wireless Edge Learning
Wang, J., Liang, B., Dong, M., Boudreau, G., & Abou-Zeid, H. (2023) · IEEE/ACM Transactions on Networking · doi:10.1109/tnet.2023.3318474 · Cited by 12
A Novel GA-BP Neural Network for Wireless Diagnosis of Rolling Bearing
Zhu, Z., Xu, X., Li, L., Dai, Y., & Meng, Z. (2022) · Journal of Circuits Systems and Computers · doi:10.1142/s0218126622501730 · Cited by 12
Prototype of deployment of Federated Learning with IoT devices
García-Santaclara, P., Vilas, A. F., & Redondo, R. P. D. (2022) · doi:10.1145/3551663.3558681 · Cited by 12
“There is a growing common view that the transition from centralized ML to distributed ML at the network edge is necessary, but largely complex, for a number of reasons [10].”
BackgroundEdge-Centric Optimization of Multi-modal ML-Driven eHealth Applications
Kanduri, A., Shahhosseini, S., Naeini, E. K., Alikhani, H., Liljeberg, P., Dutt, N. D., & Rahmani, A. M. (2023) · Springer eBooks · doi:10.1007/978-3-031-40677-5_5 · Cited by 12
Feature-Based Federated Transfer Learning: Communication Efficiency, Robustness and Privacy
Wang, F., Gursoy, M. C., & Velipasalar, S. (2024) · IEEE Transactions on Machine Learning in Communications and Networking · doi:10.1109/tmlcn.2024.3408131 · Cited by 12
“…overall dataset is imbalanced and hence the samples with certain types of output appears much more frequently than those of other outputs, it is hard for FL and FTL to distinguish this via gradient updates, and such imbalanced data distribution could significantly degrade FL performance [69], [70].”
BackgroundSingh, R., Nickhil, C., Upendar, K., Deka, S. C., & Nisha, R. (2024) · Journal of Food Measurement & Characterization · doi:10.1007/s11694-024-02999-5 · Cited by 12
“However, the adoption of such frameworks faces unique challenges in the embedded ML ecosystem, including data collection, processing, development, deployment, and monitoring [32].”
Compact and Stable Memristive Visual Geometry Group Neural Network
Ran, H., Wen, S., Li, Q., Cao, Y., Shi, K., & Huang, T. (2021) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2021.3104860 · Cited by 12
Mobile Edge Computing for Content Distribution and Mobility Support in Smart Cities
do Prado, P. F., Peixoto, M. L. M., Araújo, M. C., Gama, E. S., Gonçalves, D. M., Silva, M. V. S., Immich, R., Madeira, E. R. M., & Bittencourt, L. F. (2021) · Springer eBooks · doi:10.1007/978-3-030-69893-5_19 · Cited by 12
Client 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
“Several challenges arise when implementing FL in practice using wireless communication networks [3].”
BackgroundTraining Time Minimization in Quantized Federated Edge Learning under Bandwidth Constraint
Liu, P., Jiang, J., Zhu, G., Cheng, L., Jiang, W., Luo, W., Du, Y., & Wang, Z. (2022) · IEEE Wireless Communications and Networking Conference · doi:10.1109/wcnc51071.2022.9771723 · Cited by 12
Rate-Constrained Remote Contextual Bandits
Pase, F., Gündüz, D., & Zorzi, M. (2022) · IEEE Journal on Selected Areas in Information Theory · doi:10.1109/jsait.2022.3231459 · Cited by 12
“This demands the careful design of the distributed learning systems [3], [4], that are foreseen to produce and process most of the data in future communications networks.”
BackgroundCho, M., Lai, L., & Xu, W. (2021) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2021.3078495 · Cited by 12
“having these challenges, where machine learning process is performed through multiple decentralized devices having local data over wireless communication networks without sharing their raw data with others [5, 6].”
BackgroundRobust Federated Learning for Heterogeneous Clients and Unreliable Communications
Wang, R., Yang, L., Tang, T., Yang, B., & Wu, D. (2024) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2024.3401395 · Cited by 11
“The wireless influencing factors that affect the federated learning process were comprehensively discussed in [24].”
BackgroundHasan, M. Y., & Kadhim, D. J. (2022) · Indonesian Journal of Electrical Engineering and Computer Science · doi:10.11591/ijeecs.v25.i1.pp68-78 · Cited by 11
Dynamic Resource Allocation for Wireless Edge Machine Learning with Latency And Accuracy Guarantees
Merluzzi, M., Lorenzo, P. D., & Barbarossa, S. (2020) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp40776.2020.9052927 · Cited by 11
“The interested reader can refer to [8] for an introduction to edge machine learning, where several possible tradeoffs to be explored are presented.”
BackgroundArtificial intelligence techniques over the fifth generation mobile networks: a review
Hassan, A. N., Al-Chlaihawi, S. J., & Khekan, A. R. (2021) · Indonesian Journal of Electrical Engineering and Computer Science · doi:10.11591/ijeecs.v24.i1.pp317-328 · Cited by 11
Federated Mimic Learning for Privacy Preserving Intrusion DetectionPreprint
Al-Marri, N. A. A.-A., Çiftler, B. S., & Abdallah, M. M. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2012.06974 · Cited by 11
“Hence, FL is partially privacy-preserving and communication-wise efficient since it avoids the transmission of vast amounts of data [5], [6].”
BackgroundTowards Real-time Learning for Edge-Cloud Continuum with Vehicular Computing
Peltonen, E., Sojan, A., & Päivärinta, T. (2021) · IEEE World Forum on Internet of Things (WF-IoT) · doi:10.1109/wf-iot51360.2021.9595628 · Cited by 10
“Large-scale sensor data analytics and ML/AI solutions currently rely on novel algorithmic innovations, especially federated learning [31], [32] and other highly popular technologies [33], [34], [35].”
BackgroundUltra-Reliable and Low-Latency Communications in 6G: Challenges, Solutions, and Future Directions
She, C., & Li, Y. (2023) · Signals and communication technology · doi:10.1007/978-3-031-37920-8_24 · Cited by 10
FedADP: Communication-Efficient by Model Pruning for Federated Learning
Liu, H., Shi, Y., Su, Z., Zhang, K., Wang, X., Yan, Z., & Kong, F. (2023) · Globecom · doi:10.1109/globecom54140.2023.10437081 · Cited by 10
Contract-Based Incentive Mechanism for Federated Learning in Edge Computing System
Yu, L., Chang, Z., & Zhao, Z. (2024) · IEEE Wireless Communications and Networking Conference · doi:10.1109/wcnc57260.2024.10570831 · Cited by 10
“…train an FL algorithm over a wireless network, MUs need to transmit local model parameters over wireless channels which can introduce training and transmission errors, due to the wireless resource limitations (e.g., in terms of bandwidth and power) and the natural instability of wireless links [4].”
BackgroundAhmadi, A., Høst-Madsen, A., & Xiong, Z. (2024) · IEEE Symposium on Computers and Communications (ISCC) · doi:10.1109/iscc61673.2024.10733563 · Cited by 10
“The packet size of each IoT devices is selected randomly from a uniform distribution with a range [1, 10] kbits.”
BackgroundLin, S. (2021) · Mobile Information Systems · doi:10.1155/2021/3443189 · Cited by 10
“In recent years, the wireless sharing network occupies an important position in the national household and various enterprises, and it brings convenience to people’s life and work [12].”
BackgroundTactical Edge IoT in Defense and National Security
Fraga‐Lamas, P., & Fernández‐Caramés, T. M. (2022) · Wiley eBooks · doi:10.1002/9781119892199.ch20 · Cited by 10
Energy-Efficient Classification at the Wireless Edge with Reliability Guarantees
Merluzzi, M., Battiloro, C., Lorenzo, P. D., & Strinati, E. C. (2022) · IEEE International Conference on Communications workshops · doi:10.1109/iccworkshops53468.2022.9814659 · Cited by 9
A Survey: Network Feature Measurement Based on Machine Learning
Sun, M., He, B., Li, R., Li, J., & Zhang, X. (2023) · Applied Sciences · doi:10.3390/app13042551 · Cited by 9
Machine Learning in Future Wireless Communications: Innovations, Applications, and Open Challenges
Fatima, S., & Kondamuri, S. R. (2025) · Wireless Personal Communications · doi:10.1007/s11277-025-11770-y · Cited by 10
“However, this approach requires efficient communication protocols to manage data exchange between devices and edge servers [34].”
MotiLearn: Contract-Based Incentive Mechanism for Heterogeneous Edge Collaborative Training
Wang, Q., Guo, S., Liu, G., Yang, L., & Pan, C. (2022) · IEEE Transactions on Network Science and Engineering · doi:10.1109/tnse.2022.3173148 · Cited by 9
Data-Representation Aware Resource Scheduling for Edge Intelligence
Zeng, Z., Liu, Y., & Tang, W. (2022) · IEEE Transactions on Vehicular Technology · doi:10.1109/tvt.2022.3195571 · Cited by 9
“Towardthisregard,manyeffortshavebeenmadetomakeedge intelligencemorecommunication-efficient(seethesurvey[6]andref-erencestherein),likedataquantization,compression,partition,coding and so on.”
Multi-Resource Allocation for On-Device Distributed Federated Learning Systems
Gao, Y., Ye, Z., Yu, H., Xiong, Z., Xiao, Y., & Niyato, D. T. (2022) · Globecom · doi:10.1109/globecom48099.2022.10000935 · Cited by 9
“Meanwhile, with the increasingly advanced sensors, computing, and communication capabilities equipped to intelligent mobile devices, performing training directly at the edge device is a promising way out of this gridlock, which is commonly referred as edge ML [3].”
Background
Metadata
- Authors
- Park, J., Samarakoon, S., Bennis, M., & Debbah, M.
- Affiliations
- University of Oulu (FI)
- Université Paris-Saclay (FR)
- Venue
- Proceedings of the IEEE
- Published
- 2019-11
- Volume
- 107
- Issue
- 11
- Pages
- 2204-2239
- ISSN
- 0018-9219 (ISSN-L)
- 1558-2256
- OpenAlex
- W2904648777
- Cited by
- 92
- References
- 224
- Citations by year
- 2026: 8 · 2025: 2 · 2024: 6 · 2023: 10 · 2022: 13 · 2021: 8 · 2020: 29 · 2019: 16
- FWCI
- 9.03
- Type
- journal article
- Language
- en
- Open access
- green
- arxiv.org
- Licence
- Locations
- Proceedings of the IEEE (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- Figshare (submittedVersion, cc-by)
- arXiv (Cornell University)
- Funders
- Academy of Finland (294128)
- 6Genesis Flagship (318927)
- Kvantum Institute Strategic Project SAFARI
- Academy of Finland through the MISSION Project (319759)
- Artificial Intelligence for Mobile Wireless Systems AIMS project at the University of Oulu
- Topics
- Physical Sciences › Computer Science › Computer Networks and Communications › IoT and Edge/Fog Computing
- Physical Sciences › Computer Science › Artificial Intelligence › Privacy-Preserving Technologies in Data
- Physical Sciences › Computer Science › Computer Networks and Communications › Distributed Sensor Networks and Detection Algorithms
- Keywords
- edge machine learning, wireless networks, distributed learning, resource-constrained edge devices, 5G, federated learning, low-latency inference, autonomous systems, unmanned aerial vehicles, virtual and augmented reality
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.