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 101–200 of 702
Deep Learning at the Mobile Edge: Opportunities for 5G Networks
McClellan, M., Cervelló-Pastor, C., & Sallent, S. (2020) · Applied Sciences · doi:10.3390/app10144735 · Cited by 121
“Previous surveys have focused on categorizing and evaluating various aspects of edge computing machine learning algorithms applied to the edge of the network [6], on creating a taxonomy and description of challenges in distributed network computing [7], and the integration of mobile edge computing in 5G [8].”
BackgroundDistributed Artificial Intelligence-as-a-Service (DAIaaS) for Smarter IoE and 6G Environments
Janbi, N., Katib, I., Albeshri, A., & Mehmood, R. (2020) · Sensors · doi:10.3390/s20205796 · Cited by 102
6G White Paper on Edge Intelligence
Peltonen, E., Bennis, M., Capobianco, M., Debbah, M., Ding, A., Gil-Castiñeira, F., Jurmu, M., Karvonen, T., Kelanti, M., Kliks, A., et al. (2020) · arXiv · doi:10.48550/arxiv.2004.14850 · Cited by 118
“UTEDLY TRAINED ALGORITHMS Towards realizing edge intelligence, the training procedure directly affects the majority of the end-to-end latencies, the inference reliability, and the overall scalability [8]. While a handful of applications may allow traditional centralized artificial intelligence model training and downl…”
BackgroundSelf-citationHigh-Dimensional Stochastic Gradient Quantization for Communication-Efficient Edge Learning
Du, Y., Yang, S., & Huang, K. (2020) · IEEE Transactions on Signal Processing · doi:10.1109/tsp.2020.2983166 · Cited by 116
“edge [1].”
Improving TCP Performance Over WiFi for Internet of Vehicles: A Federated Learning Approach
Pokhrel, S. R., & Choi, J. (2020) · IEEE Transactions on Vehicular Technology · doi:10.1109/tvt.2020.2984369 · Cited by 108
“Furthermore, the existing models [5]–[9] only provide expressions for computing their performance when loss and delay are exogenous quantities.”
BackgroundThe Internet of Audio Things: State of the Art, Vision, and Challenges
Turchet, L., Fazekas, G., Lagrange, M., Shokri‐Ghadikolaei, H., & Fischione, C. (2020) · IEEE Internet of Things Journal · doi:10.1109/jiot.2020.2997047 · Cited by 114
“…resources prohibit the use of heavy training and/or inference algorithms [147], short in-device storage challenges the deployment of heavy pretrained models [148], and low bandwidth links and real-time nature of the audio signals hinders the use of traditional cloud-based inference [149], [150].”
BackgroundAbbas, Q., & Alsheddy, A. (2020) · Sensors · doi:10.3390/s21010056 · Cited by 111
“Currently, mobile-edge computing (MEC) [208] on 5G networks handle many images and video processing in a real-time environment to a strict and faster response.”
BackgroundOptimizing AI Service Placement and Resource Allocation in Mobile Edge Intelligence Systems
Lin, Z., Bi, S., & Zhang, Y. A. (2021) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2021.3081991 · Cited by 112
The Internet of Sounds: Convergent Trends, Insights, and Future Directions
Turchet, L., Lagrange, M., Rottondi, C., Fazekas, G., Peters, N., Østergaard, J., Font, F., Bäckström, T., & Fischione, C. (2023) · IEEE Internet of Things Journal · doi:10.1109/jiot.2023.3253602 · Cited by 111
“2) The second challenge for ML in IoS is that the data sets are distributed and must be connected over a public communication network, including via Edge computing [281], [284].”
End-to-End Intent-Based Networking
Velasco, L., Signorelli, M., de Dios, Ó. G., Papagianni, C., Bifulco, R., Olmos, J. J. V., Pryor, S., Carrozzo, G., Schulz-Zander, J., Bennis, M., et al. (2021) · IEEE Communications Magazine · doi:10.1109/mcom.101.2100141 · Cited by 105
“1 Intelligent Applications The possibility to move intelligence to the edge has sparked a groundswell interest in distributed on-device ML, in which training data is stored across many geographically dispersed nodes [1].”
BackgroundSelf-citationA Survey on Semantic Communications for Intelligent Wireless Networks
Iyer, S., Khanai, R., Torse, D. A., Pandya, R. J., Rabie, K. M., Pai, K., Khan, W. U., & Fadlullah, Z. M. (2022) · Wireless Personal Communications · doi:10.1007/s11277-022-10111-7 · Cited by 104
“The authors in [59] have addressed this issue for the case of edge learning in which the tools for learning are much closer to the user equipment to provision the applications with acute constraints of delay.”
BackgroundReview of the State of the Art on Adaptive Protection for Microgrids Based on Communications
Gutiérrez-Rojas, D., Nardelli, P. H. J., Mendes, G., & Popovski, P. (2020) · IEEE Transactions on Industrial Informatics · doi:10.1109/tii.2020.3006845 · Cited by 101
“filters [126] and edge intelligence, are expected to take place in the upcoming years[127].”
BackgroundBenchmark Analysis of YOLO Performance on Edge Intelligence Devices
Feng, H., Mu, G., Zhong, S., Zhang, P., & Yuan, T. (2022) · Cryptography · doi:10.3390/cryptography6020016 · Cited by 100
Dhinesh Kumar, R., & Chavhan, S. (2022) · Sustainable Energy Technologies and Assessments · doi:10.1016/j.seta.2022.102666 · Cited by 100
Federated Knowledge DistillationPreprint
Seo, H., Park, J., Oh, S., Bennis, M., & Kim, S. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2011.02367 · Cited by 98
Self-citationFederated Learning for Hybrid Beamforming in mm-Wave Massive MIMO
Elbir, A. M., & Çöleri, S. (2020) · IEEE Communications Letters · doi:10.1109/lcomm.2020.3019312 · Cited by 96
“To alleviate the transmission overhead problem, federated learning (FL) strategies are recently introduced, especially for edge computing applications [6].”
BackgroundBhat, S. A., Sofi, I. B., & Chi, C. (2020) · IEEE Access · doi:10.1109/access.2020.3037108 · Cited by 97
Machine Intelligence at the Edge With Learning Centric Power Allocation
Wang, S., Wu, Y., Xia, M., Wang, R., & Poor, H. V. (2020) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2020.3010522 · Cited by 95
The Challenges of Leveraging Threat Intelligence to Stop Data Breaches
Ibrahim, A. S., Thiruvady, D., Schneider, J.-G., & Abdelrazek, M. (2020) · Frontiers in Computer Science · doi:10.3389/fcomp.2020.00036 · Cited by 93
A Survey on Goal-Oriented Semantic Communication: Techniques, Challenges, and Future Directions
Getu, T. M., Kaddoum, G., & Bennis, M. (2024) · IEEE Access · doi:10.1109/access.2024.3381967 · Cited by 96
Self-citationData-Importance Aware User Scheduling for Communication-Efficient Edge Machine Learning
Liu, D., Zhu, G., Zhang, J., & Huang, K. (2020) · IEEE Transactions on Cognitive Communications and Networking · doi:10.1109/tccn.2020.2999606 · Cited by 92
Machine Learning Threatens 5G Security
Suomalainen, J., Juhola, A., Shahabuddin, S., Mämmelä, A., & Ahmad, I. (2020) · IEEE Access · doi:10.1109/access.2020.3031966 · Cited by 91
“For example, ML will be applied in the access network to increase spectral efficiency or for other intelligent use of radio resources [27]; in the edge near the access network to intelligently serve latency-critical services by providing higher resources in the edge [28] and IoT [29]; in the backhaul or transport netw…”
MethodPyramid: Enabling Hierarchical Neural Networks with Edge Computing
He, Q., Dong, Z., Chen, F., Deng, S., Liang, W., & Yang, Y. J. (2022) · International World Wide Web Conference (WWW) · doi:10.1145/3485447.3511990 · Cited by 90
“However, it is difficult and often impossible to obtain a high inference accuracy by training an ML model on an individual end-device due to the low quantity and/or diversity of training data [42].”
BackgroundYang, S., Ding, G., Chen, Z., & Yang, J. S. (2025) · IEEE Access · doi:10.1109/access.2025.3633290 · Cited by 89
“Park et al. [49] proposed Placeto, which uses RL for container placement in edge computing environments.”
Adaptive Configuration for Heterogeneous Participants in Decentralized Federated Learning
Liao, Y., Xu, Y., Xu, H., Wang, L., & Qian, C. (2023) · IEEE Conference on Computer Communications (INFOCOM) · doi:10.1109/infocom53939.2023.10228945 · Cited by 88
Importance-Aware Data Selection and Resource Allocation in Federated Edge Learning System
He, Y., Ren, J., Yu, G., & Yuan, J. (2020) · IEEE Transactions on Vehicular Technology · doi:10.1109/tvt.2020.3015268 · Cited by 87
“To address both issues, federated edge learning (FEEL) [6]–[8] has been recently proposed by combining federated learning (FL) [9], a specific distributed training framework, with the mobile edge computing (MEC) [10].”
AI in 6G: Energy-Efficient Distributed Machine Learning for Multilayer Heterogeneous Networks
Hossain, M. A., Hossain, A. R., & Ansari, N. (2022) · IEEE Network · doi:10.1109/mnet.104.2100422 · Cited by 87
“…of a network with AI but rather making AI a core basis of a network because AI will be instrumental in enhancing different network functions for 6G such as resource utilization, estimating channel conditions, optimizing network slicing, enhancing encryption and security, and more [9]-[11].”
BackgroundFederated Learning under Channel Uncertainty: Joint Client Scheduling and Resource Allocation
Wadu, M. M., Samarakoon, S., & Bennis, M. (2020) · IEEE Wireless Communications and Networking Conference · doi:10.1109/wcnc45663.2020.9120649 · Cited by 86
“The proliferation of a new breed of autonomous devices and mission-critical applications sparked a huge interest in machine learning at the network edge, coined Edge ML [1].”
BackgroundSelf-citationJoint Optimization of Communications and Federated Learning Over the Air
Fan, X., Wang, Y., Huo, Y., & Tian, Z. (2021) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2021.3130111 · Cited by 86
Intelligent Edge: Leveraging Deep Imitation Learning for Mobile Edge Computation Offloading
Yu, S., Chen, X., Yang, L., Wu, D., Bennis, M., & Zhang, J. (2020) · IEEE Wireless Communications · doi:10.1109/mwc.001.1900232 · Cited by 81
“org/technologies-clusters/technologies/ multi-access-edge-computing [2] J.”
BackgroundSelf-citationVehicular Cooperative Perception Through Action Branching and Federated Reinforcement Learning
Abdel-Aziz, M. K., Perfecto, C., Samarakoon, S., Bennis, M., & Saad, W. (2021) · IEEE Transactions on Communications · doi:10.1109/tcomm.2021.3126650 · Cited by 85
“Moreover, in order to enhance the training process, we propose the use of federated RL [27]–[29].”
Self-citationFlexiFed: Personalized Federated Learning for Edge Clients with Heterogeneous Model Architectures
Wang, K., He, Q., Chen, F., Chen, C., Huang, F., Jin, H., & Yang, Y. J. (2023) · International World Wide Web Conference (WWW) · doi:10.1145/3543507.3583347 · Cited by 85
“However, it raises the performance issue because such ML models often suffer from low model accuracy due to the lack of high-quality training data from diverse sources [23, 43].”
BackgroundYang, X., Yu, X., Huang, H., & Zhu, H. (2019) · IEEE Access · doi:10.1109/access.2019.2936435 · Cited by 85
“Nowadays, in the direction of deep learning, wireless networks are becoming smarter [31], [32].”
BackgroundRobust Blockchained Federated Learning with Model Validation and Proof-of-Stake Inspired Consensus
Chen, H., Asif, S. A., Park, J., Shen, C.-C., & Bennis, M. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2101.03300 · Cited by 37
“In search of the sheer amount of fresh data, networked edge devices, such as phones, cameras, and Internet of Things (IoT), from which the freshest data are generated have recently attracted significant attention (Park et al. 2019a,b, 2020).”
BackgroundSelf-citationA Neural Network-Based On-device Learning Anomaly Detector for Edge Devices
Tsukada, M., Kondo, M., & Matsutani, H. (2020) · IEEE Transactions on Computers · doi:10.1109/tc.2020.2973631 · Cited by 81
“Hence, the power consumption of ONLAD Core itself is even lower than 3.1 W. studied federated learning approaches with edge devices on wireless sensor networks [40], [41].”
InfluentialMassive Autonomous UAV Path Planning: A Neural Network Based Mean-Field Game Theoretic Approach
Shiri, H., Park, J., & Bennis, M. (2019) · Globecom · doi:10.1109/globecom38437.2019.9013181 · Cited by 80
“Collaborative HJB and FPK model training across UAVs via federated learning frameworks [17] could thus be an interesting topic for future work.”
BackgroundSelf-citationBandwidth Allocation for Multiple Federated Learning Services in Wireless Edge Networks
Xu, J., Wang, H., & Chen, L. (2021) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2021.3113346 · Cited by 79
“With the help of multi-access edge computing (MEC) servers, ML models can be quickly trained/updated using this data to adapt to the changing environment without moving the data to the remote cloud data center, which is envisioned in intelligent next-generation communication systems [1], [2].”
MethodVirtual Reality Solutions Employing Artificial Intelligence Methods: A Systematic Literature Review
de Oliveira, T. R., Rodrigues, B. B., 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., et al. (2022) · ACM Computing Surveys · doi:10.1145/3565020 · Cited by 81
Efficient Federated Meta-Learning Over Multi-Access Wireless Networks
Yue, S., Ren, J., Xin, J., Zhang, D., Zhang, Y., & Zhuang, W. (2022) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2022.3143259 · Cited by 77
Delay Minimization for Federated Learning Over Wireless Communication NetworksPreprint
Yang, Z., Chen, M., Saad, W., Hong, C. S., Shikh‐Bahaei, M., Poor, H. V., & Cui, S. R. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2007.03462 · Cited by 77
“To this end, distributed edge learning approaches, such as federated learning (FL), were proposed (Saad et al., to appear, 2020; Park et al., 2019; Chen et al., 2020; Samarakoon et al., 2018; G¨und¨uz et al., 2019; Chen et al., 2019b).”
MethodFederated Learning for RSS Fingerprint-based Localization: A Privacy-Preserving Crowdsourcing Method
Ciftler, B. S., Albaseer, A., Lasla, N., & Abdallah, M. (2020) · 2020 International Wireless Communications and Mobile Computing (IWCMC) · doi:10.1109/iwcmc48107.2020.9148111 · Cited by 76
“Hence, federated learning is also communication-wise efficient since it avoids the transmission of vast amounts of data [7].”
BackgroundQ-GADMM: Quantized Group ADMM for Communication Efficient Decentralized Machine Learning
Elgabli, A., Park, J., Bedi, A. S., Issaid, C. B., Bennis, M., & Aggarwal, V. (2021) · IEEE Transactions on Communications · doi:10.1109/tcomm.2020.3026398 · Cited by 51
“For a large scale network, the communication cost often becomes dominant compared to the computation time, calling for communication efficient distributed optimization [5], [39]–[41].”
Self-citationOh, S., Park, J., Vepakomma, P., Baek, S., Raskar, R., Bennis, M., & Kim, S. (2022) · International World Wide Web Conference (WWW) · doi:10.1145/3485447.3512153 · Cited by 76
“In this respect, the indispensable elements are the utilization of massive amounts of data and the provision of large-scale parallel computing power [5, 12, 20, 28].”
BackgroundSelf-citationEnabling All In-Edge Deep Learning: A Literature Review
Joshi, P., Hasanuzzaman, M., Thapa, C., Afli, H., & Scully, T. (2023) · IEEE Access · doi:10.1109/access.2023.3234761 · Cited by 77
“Unlike prior surveys [13], [15], [16], [17], [18],”
SLO-Aware Inference Scheduler for Heterogeneous Processors in Edge Platforms
Seo, W., Cha, S.-H., Kim, Y., Huh, J., & Park, J. (2021) · ACM Transactions on Architecture and Code Optimization · doi:10.1145/3460352 · Cited by 75
“Built upon such outstanding advances, the industry is now moving toward integrating the ML algorithms into various types of real-world applications and deploying the applications on the edge platforms [4, 22, 23, 29, 33, 35, 38, 47, 63, 65].”
BackgroundSelf-citationMix2FLD: Downlink Federated Learning After Uplink Federated Distillation With Two-Way Mixup
Oh, S., Park, J., Jeong, E., Kim, H., Bennis, M., & Kim, S. (2020) · IEEE Communications Letters · doi:10.1109/lcomm.2020.3003693 · Cited by 74
“INTRODUCTION USER-GENERATED local data is essential in training machine learning (ML) models for mission-critical applications, but exchanging data may violate privacy and induce huge communication overhead [1].”
BackgroundMethodSelf-citationAttention-based Reinforcement Learning for Real-Time UAV Semantic Communication
Yun, W. J., Lim, B., Jung, S., Ko, Y., Park, J., Kim, J., & Bennis, M. (2021) · International Symposium on Wireless Communication Systems · doi:10.1109/iswcs49558.2021.9562230 · Cited by 70
Self-citationGuo, S., Lu, B., Wen, M., Dang, S., & Saeed, N. (2022) · IEEE Communications Standards Magazine · doi:10.1109/mcomstd.0001.2100041 · Cited by 70
Lightweight Blockchain-Empowered Secure and Efficient Federated Edge Learning
Jin, R. M., Hu, J. W., Min, G., & Mills, J. (2023) · IEEE Transactions on Computers · doi:10.1109/tc.2023.3293731 · Cited by 70
“Nowadays, networked edge devices, such as smartphones, vehicles, and sensors contain large amounts of valuable data [1].”
Toward Multiple Federated Learning Services Resource Sharing in Mobile Edge Networks
Nguyen, M. N. H., Tran, N. H. K., Tun, Y. K., Han, Z., & Hong, C. S. (2021) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2021.3085979 · Cited by 69
“edge ML applications for the vision of truly intelligent nextgeneration communication systems in 6G [1].”
BackgroundFederated Mimic Learning for Privacy Preserving Intrusion Detection
Al-Athba Al-Marri, N. A., Ciftler, B. S., & Abdallah, M. M. (2020) · 2020 IEEE International Black Sea Conference on Communications and Networking (BlackSeaCom) · doi:10.1109/blackseacom48709.2020.9234959 · Cited by 68
“Hence, FL is partially privacy-preserving and communication-wise efficient since it avoids the transmission of vast amounts of data [5], [6].”
BackgroundTowards Cooperative Data Rate Prediction for Future Mobile and Vehicular 6G Networks
Sliwa, B., Falkenberg, R., & Wietfeld, C. (2020) · Joint European Conference on Networks and Communications & 6G Summit (EuCNC/6G Summit) · doi:10.1109/6gsummit49458.2020.9083767 · Cited by 68
“While the ongoing standardization efforts for 5G mainly target network-side intelligence, different studies have demonstrated that edge-based [4] and UE-based optimization is not only able to improve the end user experience, but also contributes to improving the intra-cell coexistence of different devices.”
BackgroundCommunication-Efficient Large-Scale Distributed Deep Learning: A Comprehensive SurveyPreprint
Liang, F., Zhang, Z., Lu, H., Leung, V. C. M., Guo, Y., & Hu, X. (2024) · arXiv (Cornell University) · doi:10.48550/arxiv.2404.06114 · Cited by 67
“Several surveys address the communication challenges in specific distributed DL paradigms, including cloud-edge-based [33]–[38], P2P-based [39], and hybrid [40], [41] paradigms.”
BackgroundWireless Edge Machine Learning: Resource Allocation and Trade-Offs
Merluzzi, M., Lorenzo, P. D., & Barbarossa, S. (2021) · IEEE Access · doi:10.1109/access.2021.3066559 · Cited by 66
“A first general introduction to EML can be found in [20], where the authors present several possible trade-offs.”
BackgroundEnergy-Efficient Processing and Robust Wireless Cooperative Transmission for Edge Inference
Yang, K., Shi, Y., Yu, W., & Ding, Z. (2020) · IEEE Internet of Things Journal · doi:10.1109/jiot.2020.2979523 · Cited by 65
“However, since mobile devices are usually equipped with limited computation power, storage and energy [2], it is usually infeasible to deploy deep learning models, i.”
InfluentialVertical Federated Edge Learning With Distributed Integrated Sensing and Communication
Liu, P., Zhu, G., Jiang, W., Luo, W., Xu, J., & Cui, S. R. (2022) · IEEE Communications Letters · doi:10.1109/lcomm.2022.3181612 · Cited by 64
“abundant intelligent applications such as auto-driving, smart home, and object/activity recognition [1].”
BackgroundMethodTask-Oriented Communication Design in Cyber-Physical Systems: A Survey on Theory and Applications
Mostaani, A., Vu, T. X., Sharma, S. K., Nguyen, V., Liao, Q., & Chatzinotas, S. (2022) · IEEE Access · doi:10.1109/access.2022.3231039 · Cited by 64
“FL is capable of addressing many challenges in implementing ML over networks [182], [183].”
BackgroundRemote UAV Online Path Planning via Neural Network-Based Opportunistic Control
Shiri, H., Park, J., & Bennis, M. (2020) · IEEE Wireless Communications Letters · doi:10.1109/lwc.2020.2973624 · Cited by 64
“Extending this framework to incorporating deep NN architectures [19] with multiple UAVs could be an interesting topics for future research.”
BackgroundSelf-citationSelf Organizing Federated Learning Over Wireless Networks: A Socially Aware Clustering Approach
Khan, L. U., Alsenwi, M., Han, Z., & Hong, C. S. (2020) · 2020 International Conference on Information Networking (ICOIN) · doi:10.1109/icoin48656.2020.9016505 · Cited by 59
Ahmad, I., Shahabuddin, S., Sauter, T., Harjula, E., Kumar, T., Meisel, M., Juntti, M. J., & Ylianttila, M. (2020) · IEEE Industrial Electronics Magazine · doi:10.1109/mie.2020.2979272 · Cited by 64
“If resources are available in highend servers in centralized cloud systems, latencycritical applications will be challenged by communication latency, besides other factors [7].”
BackgroundA Hybrid Architecture for Federated and Centralized Learning
Elbir, A. M., Çöleri, S., Papazafeiropoulos, A., Kourtessis, P., & Chatzinotas, S. (2022) · IEEE Transactions on Cognitive Communications and Networking · doi:10.1109/tccn.2022.3181032 · Cited by 63
“Many of the ML tasks are based on centralized learning (CL) algorithms, which train powerful ANNs at a parameter server (PS) [2], [10], [11].”
BackgroundFine-Grained Data Selection for Improved Energy Efficiency of Federated Edge Learning
Albaseer, A. M., Abdallah, M. M., Al‐Fuqaha, A., & Erbad, A. (2021) · IEEE Transactions on Network Science and Engineering · doi:10.1109/tnse.2021.3100805 · Cited by 63
“FEEL can be seen as a cutting-edge collaborative machine learning (ML) technique for future IoT and edge systems [11]–[14].”
BackgroundGirgis, A. M., Park, J., Bennis, M., & Debbah, M. A. (2021) · IEEE Transactions on Communications · doi:10.1109/tcomm.2021.3099156 · Cited by 62
Self-citationRevisiting Analog Over-the-Air Machine Learning: The Blessing and Curse of Interference
Yang, H. H., Chen, Z., Quek, T. Q. S., & Poor, H. V. (2021) · IEEE Journal of Selected Topics in Signal Processing · doi:10.1109/jstsp.2021.3139231 · Cited by 62
Lyapunov-Based Optimization of Edge Resources for Energy-Efficient Adaptive Federated Learning
Battiloro, C., Lorenzo, P. D., Merluzzi, M., & Barbarossa, S. (2022) · IEEE Transactions on Green Communications and Networking · doi:10.1109/tgcn.2022.3186879 · Cited by 62
“As such, enabling edge ML introduces novel research problems in terms of jointly optimizing inference, training, communication, computation, and control under end-to-end latency, reliability, and learning performance requirements [7]–[10].”
BackgroundKokkonen, H., Lovén, L., Motlagh, N. H., Kumar, A., Partala, J., Nguyen, T., Pujol, V. C., Kostakos, P., Leppänen, T., González-Gil, A., et al. (2022) · arXiv (Cornell University) · doi:10.48550/arxiv.2205.01423 · Cited by 63
“[14] provide a very comprehensive survey of communicationefficient model building on edge, focusing on introducing technical and theoretical enablers for accurate and low latency model training and inference.”
BackgroundMethodInfluentialSelf-citationRelay-Assisted Cooperative Federated Learning
Lin, Z., Liu, H., & Zhang, Y. A. (2022) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2022.3155596 · Cited by 61
Intelligent Communications for Tactile Internet in 6G: Requirements, Technologies, and Challenges
Hou, Z., She, C., Li, Y., Niyato, D. T., Döhler, M., & Vucetic, B. S. (2021) · IEEE Communications Magazine · doi:10.1109/mcom.006.2100227 · Cited by 63
Content-based Vehicle Selection and Resource Allocation for Federated Learning in IoV
Wang, S., Liu, F., & Xia, H. (2021) · IEEE Wireless Communications and Networking Conference Workshops (WCNCW) · doi:10.1109/wcncw49093.2021.9419986 · Cited by 61
“Moreover, since the state of the wireless channel is unstable, the probability of packet error in the transmission process greatly increases and the limited bandwidth will limit the number of vehicle clients participating in federated learning [12].”
BackgroundClient Selection Approach in Support of Clustered Federated Learning over Wireless Edge Networks
Albaseer, A. M., Abdallah, M. M., Al‐Fuqaha, A., & Erbad, A. (2021) · Globecom · doi:10.1109/globecom46510.2021.9685938 · Cited by 61
GADMM: Fast and Communication Efficient Framework for Distributed Machine Learning
Elgabli, A., Park, J., Bedi, A. S., Bennis, M., & Aggarwal, V. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1909.00047 · Cited by 42
Self-citationFedADC: Accelerated Federated Learning with Drift Control
Özfatura, E., Özfatura, K., & Gündüz, D. (2021) · IEEE International Symposium on Information Theory (ISIT) · doi:10.1109/isit45174.2021.9517850 · Cited by 60
Distributed Learning for Wireless Communications: Methods, Applications and Challenges
Qian, L., Yang, P., Xiao, M., Dobre, O. A., Renzo, M. D., Li, J., Han, Z., Qin, Y., & Zhao, J. (2022) · IEEE Journal of Selected Topics in Signal Processing · doi:10.1109/jstsp.2022.3156756 · Cited by 60
Network Digital Twin Toward Networking, Telecommunications, and Traffic Engineering: A Survey
Poorzare, R., Kanellopoulos, D. N., Sharma, V. K., Dalapati, P., & Waldhorst, O. P. (2025) · IEEE Access · doi:10.1109/access.2025.3531947 · Cited by 60
Wireless for Machine Learning: A Survey
Hellström, H., da Silva, J. M. B., Amiri, M. M., Chen, M., Fodor, V., Poor, H. V., & Fischione, C. (2022) · Foundations and Trends® in Signal Processing · doi:10.1561/2000000114 · Cited by 59
Inexact-ADMM Based Federated Meta-Learning for Fast and Continual Edge Learning
Yue, S., Ren, J., Xin, J., Lin, S., & Zhang, J. (2021) · ACM International Symposium on Theory, Algorithmic Foundations, and Protocol Design for Mobile Networks and Mobile Computing (MobiHoc) · doi:10.1145/3466772.3467038 · Cited by 59
“Clearly, owing to the limitation of bandwidth and the concern for privacy leakage, the conventional cloud intelligence/computing would not work well in this case, which facilitates a new research area, namely edge intelligence or edge AI [24, 45].”
BackgroundMP-FedCL: Multiprototype Federated Contrastive Learning for Edge Intelligence
Qiao, Y., Munir, M. S., Adhikary, A., Le, H. Q., Raha, A. D., Zhang, C., & Hong, C. S. (2023) · IEEE Internet of Things Journal · doi:10.1109/jiot.2023.3320250 · Cited by 58
“In this fashion, FL’s ability to protect data confidentiality and enable multiple parties to cooperatively train a model makes it a highly promising technology for the future of network intelligence [7].”
BackgroundTracing the Influence of Large Language Models across the Most Impactful Scientific Works
Petroșanu, D.-M., Pîrjan, A., & Tăbușcă, A. (2023) · Electronics · doi:10.3390/electronics12244957 · Cited by 58
Chougule, S., Chaudhari, B. S., Ghorpade, S. N., & Zennaro, M. (2024) · World Electric Vehicle Journal · doi:10.3390/wevj15020039 · Cited by 58
Federated Edge Learning for 6G: Foundations, Methodologies, and Applications
Tao, M., Zhou, Y., Shi, Y., Lu, J., Cui, S. R., Lü, J., & Letaief, K. B. (2024) · Proceedings of the IEEE · doi:10.1109/jproc.2024.3509739 · Cited by 58
“enabling ubiquitous and trustworthy AI applications in wireless networks [9], [10], [11].”
Wireless for Machine Learning: A Survey
Hellström, H., da Silva, J. M. B., Amiri, M. M., Chen, M., Fodor, V., Poor, H. V., & Fischione, C. (2022) · doi:10.1561/9781638280071 · Cited by 57
Over-the-Air Federated Learning and Optimization
Zhu, J., Shi, Y., Zhou, Y., Jiang, C., Chen, W., & Letaief, K. B. (2024) · IEEE Internet of Things Journal · doi:10.1109/jiot.2024.3352280 · Cited by 56
“To this end, federated edge learning (FL) [5], [6], [7], as an emerging distributed ML framework, has been proposed to enable many edge devices to train an ML model under the coordination of an edge server without sharing their local data.”
BackgroundOptimizing Edge AI: A Comprehensive Survey on Data, Model, and System StrategiesPreprint
Wang, X., & Jia, W. (2025) · Qeios · doi:10.32388/izohch · Cited by 56
Ensemble Federated Learning With Non-IID Data in Wireless Networks
Zhao, Z., Wang, J., Hong, W., Quek, T. Q. S., Ding, Z., & Peng, M. (2023) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2023.3309376 · Cited by 55
“Recently, network intelligenti-zation is extending from the core networks to the network edge devices [5], [6], [7], which can support flexible deployment and orchestration of deep learning/machine learning models to satisfy ultimate and diverse quality-of-service (QoS) requirements of future 6G…”
BackgroundCommunications Security in Industry X: A Survey
Ahmad, I., Rodriguez, F., Kumar, T., Suomalainen, J., Jagatheesaperumal, S. K., Walter, S., Asghar, M. Z., Li, G., Papakonstantinou, N., Ylianttila, M., et al. (2024) · IEEE Open Journal of the Communications Society · doi:10.1109/ojcoms.2024.3356076 · Cited by 56
“With 5G providing computing and storage capabilities in the vicinity of InX, machine learning in the network edge [125] in InX will be facilitated by 5G [126].”
MethodCommunication-Efficient Federated Edge Learning via Optimal Probabilistic Device Scheduling
Zhang, M., Zhu, G., Wang, S., Jiang, J., Liao, Q., Zhong, C., & Cui, S. R. (2022) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2022.3166941 · Cited by 54
AI augmented edge and fog computing for Internet of Health Things (IoHT)
Rajagopal, D., & Subramanian, P. K. T. (2025) · PeerJ Computer Science · doi:10.7717/peerj-cs.2431 · Cited by 54
Wireless Data Acquisition for Edge Learning: Data-Importance Aware Retransmission
Liu, D., Zhu, G., Zeng, Q., Zhang, J., & Huang, K. (2020) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2020.3024980 · Cited by 54
Federated Learning-based Driver Activity Recognition for Edge Devices
Doshi, K., & Yilmaz, Y. (2022) · 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw56347.2022.00377 · Cited by 53
“Fed-erated Learning (FL) is offered as a distributed model training method that does not communicate raw data, therefore maintaining data privacy and saving communication band-width [16, 26, 36, 37].”
BackgroundDistributed Machine Learning through Heterogeneous Edge Systems
Hu, H., Wang, D., & Wu, C. (2020) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v34i05.6207 · Cited by 53
Over-the-Air Multi-View Pooling for Distributed Sensing
Liu, Z., Lan, Q., Kalør, A. E., Popovski, P., & Huang, K. (2023) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2023.3343086 · Cited by 52
“One is the ubiquitous deployment of artificial intelligence (AI) algorithms at the network edge, termed edge AI, to empower Internet-of-things (IoT) applications [2], [3].”
BackgroundArtificial Intelligence, Sensors and Vital Health Signs: A Review
Junaid, S. B., Imam, A. A., Shuaibu, A. N., Basri, S. B., Kumar, G., Surakat, Y. A., Balogun, A. O., Abdulkarim, M., Garba, A., Sahalu, Y., et al. (2022) · Applied Sciences · doi:10.3390/app122211475 · Cited by 51
Balasubramanian, V., Aloqaily, M., Martin Reisslein, B., & Scaglione, A. (2021) · IEEE Network · doi:10.1109/mnet.011.2100121 · Cited by 51
Strinati, E. C., Barbarossa, S., Gonzalez-Jimenez, J. L., Kténas, D., Cassiau, N., & Dehos, C. (2019) · arXiv · doi:10.48550/arxiv.1901.03239 · Cited by 49
“The exploitation of machine learning algorithms at the edge of the network has been recently suggested in [7].”
BackgroundA Detailed Relevance Analysis of Enabling Technologies for 6G Architectures
Pivoto, D. G. S., Rezende, T. T., Facina, M. S. P., Moreira, R., Silva, F. D. O., Cardoso, K. V., Corrêa, S. L., Araujo, A. V. D., Silva, W. R. S. E., Neto, H. S., et al. (2023) · IEEE Access · doi:10.1109/access.2023.3301811 · Cited by 49
“Indeed, in a survey comprising 119 articles from [22, 26-28, 30-144], we found 72 different technologies identified as 6G enablers.”
BackgroundHybrid reality development - can social responsibility concepts provide guidance?
Perko, I. (2020) · Kybernetes · doi:10.1108/k-01-2020-0061 · Cited by 49
AI-Based Network-Aware Service Function Chain Migration in 5G and Beyond Networks
Addad, R. A., Dutra, D. L. C., Taleb, T., & Flinck, H. (2021) · IEEE Transactions on Network and Service Management · doi:10.1109/tnsm.2021.3074618 · Cited by 49
Ali, Z., Abbas, Z. H., Abbas, G., Numani, A., & Bilal, M. (2021) · Computer Networks · doi:10.1016/j.comnet.2021.108356 · Cited by 49
Server-Side Local Gradient Averaging and Learning Rate Acceleration for Scalable Split LearningPreprint
Pal, S., Uniyal, M., Park, J., Vepakomma, P., Raskar, R., Bennis, M., Jeon, M., & Choi, J. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2112.05929 · Cited by 49
“like smartphones, IoT, and e-health wearables, so are inherently dispersed and often privacy-sensitive (Park et al. 2019; Smith et al. 2018).”
BackgroundSelf-citationStorage-Aware Joint User Scheduling and Bandwidth Allocation for Federated Edge Learning
Liu, S., Shen, Y., Yuan, J., Wu, C., & Yin, R. (2024) · IEEE Transactions on Cognitive Communications and Networking · doi:10.1109/tccn.2024.3451711 · Cited by 48
“To utilize these data in an efficient manner, a new computing paradigm, i.e., Edge Learning (EL), has been proposed by migrating the training procedure from central cloud servers to the edge of wireless networks [9], [10], [11].”
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.