Deep Reinforcement Learning based Resource Allocation in Low Latency Edge Computing Networks
The DOI matches its Crossref record; 15 of the 15 references it cites were found in a registry or an index.
Abstract
In this paper, we investigate strategies for the allocation of computational resources using deep reinforcement learning in mobile edge computing networks that operate with finite blocklength codes to support low latency communications. The end-to-end (E2E) reliability of the service is addressed, while both the delay violation probability and the decoding error probability are taken into account. By employing a deep reinforcement learning method, namely deep Q-learning, we design an intelligent agent at the edge computing node to develop a real-time adaptive policy for computational resource allocation for offloaded tasks of multiple users in order to improve the average E2E reliability. Via simulations, we show that under different task arrival rates, the realized policy serves to increase the task number that decreases the delay violation rate while guaranteeing an acceptable level of decoding error probability. Moreover, we show that the proposed deep reinforcement learning approach outperforms the random and equal scheduling benchmarks.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Citations
Cites
Human-level control through deep reinforcement learning
Mnih, V., Kavukcuoglu, K., Silver, D., Rusu, A. A., Veness, J., Bellemare, M. G., Graves, A., Riedmiller, M., Fidjeland, A. K., Ostrovski, G., et al. (2015) · Nature · doi:10.1038/nature14236 · Cited by 30,357
Reinforcement Learning: An Introduction
Sutton, R. S., & Barto, A. G. (1998) · IEEE Transactions on Neural Networks · doi:10.1109/tnn.1998.712192 · Cited by 27,225 · cited 1×
“This value iteration algorithm is guaranteed to converge to the optimal point with Q → Q as k →∞ [12].”
MethodReinforcement Learning: An Introduction
(2005) · IEEE Transactions on Neural Networks · doi:10.1109/tnn.2004.842673 · Cited by 24,496
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
Markov decision processes: Discrete stochastic dynamic programming
Hazeghi, K., & Puterman, M. L. (1995) · Journal of the American Statistical Association · doi:10.2307/2291177 · Cited by 8,037 · cited 1×
“ISWCS 2018 SS10 - Ultra-Reliable, Low-Latency and Low-Power Transmissions in the Era of Internet-of-Things as the solution of the Bellman equation [11]: By using the Bellman equation as an iterative update, the Q-function is estimated as: This value iteration algorithm is guaranteed to converge to…”
MethodFog computing and its role in the internet of things
Bonomi, F., Milito, R., Zhu, J., & Addepalli, S. (2012) · Proceedings of the first edition of the MCC workshop on Mobile cloud computing · doi:10.1145/2342509.2342513 · Cited by 6,244
Channel Coding Rate in the Finite Blocklength Regime
Polyanskiy, Y., Poor, H. V., & Verdu, S. (2010) · IEEE Transactions on Information Theory · doi:10.1109/tit.2010.2043769 · Cited by 3,885
Mobile Edge Computing: A Survey
Abbas, N., Zhang, Y., Taherkordi, A., & Skeie, T. (2018) · IEEE Internet of Things Journal · doi:10.1109/jiot.2017.2750180 · Cited by 2,474
A Survey on the Edge Computing for the Internet of Things
Yu, W., Liang, F., He, X., Hatcher, W. G., Lu, C., Lin, J., & Yang, X. (2018) · IEEE Access · doi:10.1109/access.2017.2778504 · Cited by 1,520
DREAM: Dynamic Resource and Task Allocation for Energy Minimization in Mobile Cloud Systems
Kwak, J., Kim, Y., Lee, J., & Chong, S. (2015) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2015.2478718 · Cited by 449
Cited by
Items 101–158 of 158
Computation Offloading in Edge Computing Based on Deep Reinforcement Learning
Li, M., Mao, N., Zheng, X., & Gadekallu, T. R. (2022) · Lecture notes in networks and systems · doi:10.1007/978-981-19-0604-6_28 · Cited by 4
Resource Allocation Technology Mobile Edge Computing System Using Blockchain Technology
Sharma, A., Rani, R. P., RaenuKolandaisamy, Abdulrahman, I. S., Sreeram, A., & Shukla, S. K. (2023) · International Conference on Advance Computing and Innovative Technologies in Engineering (ICACITE) · doi:10.1109/icacite57410.2023.10183102 · Cited by 4
Resource management for meteorological service in cloud‐edge computing: A survey
Tang, H., Zhang, X., Fu, S., Liu, X., Wu, Q., & Qi, L. (2020) · Transactions on Emerging Telecommunications Technologies · doi:10.1002/ett.3844 · Cited by 4
“93 To reinforce the average dependability of edge to edge, Yang et al 94 design a smart application at the edge of the network to evolve real-time accommodative approach for optimizing the transmit data of multiple smart devices by using a deep learning approach.”
MethodResource Allocation in Edge Computing Environment Using Deterministic Policy Gradient Algorithm
Shabariram, C. P., Shanthi, N., Ponnuswamy, P. P., Subramaniyaswamy, V., & Sathana, V. (2024) · International Conference on Advanced Computing and Communication Systems (ICACCS) · doi:10.1109/icaccs60874.2024.10717103 · Cited by 3
“Yang et al., [2] discusses the strategies for allocating computational resources in mobile edge computing networks that provide low latency communications by using finite block length codes.”
BackgroundMobile edge computing resource allocation: An operating system view
Chen, X. (2021) · Computer Networks · doi:10.1016/j.comnet.2021.107925 · Cited by 3
Zhang, Y., Tan, G., & Zhou, S. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-86130-8_35 · Cited by 3
An Adaptive Cloud Bursting Job Scheduler based on Deep Reinforcement Learning
Yasuda, S., Lee, C., & Date, S. (2021) · International Conference on High Performance Big Data and Intelligent Systems (HPBD&IS) · doi:10.1109/hpbdis53214.2021.9658447 · Cited by 3
On Meeting a Maximum Delay Constraint
Shafieirad, H., & Adve, R. (2021) · Globecom · doi:10.1109/globecom46510.2021.9685672 · Cited by 4
“The most common RL framework considered has been deep Q-learning, e.g., the increase in the number of served users by considering the delay violation rate as in [15].”
BackgroundMECaNIC: SmartNIC to Assist URLLC Processing in Multi-Access Edge Computing Platforms
Park, T., You, M., Cui, J., Jin, Y., Lee, K., & Shin, S. (2022) · IEEE International Conference on Network Protocols (ICNP) · doi:10.1109/icnp55882.2022.9940263 · Cited by 3
“Other researches on MEC focus on improving the responsiveness of the network or providing an optimized network fabric, such as optimal MEC VM / application placement in a network [47], [48], lightweight MEC service initiation for responsiveness [49]–[51], or efficient service utilization with MEC [52], [53], and overl…”
BackgroundEnergy and Time-Effective Computation Offloading for Edge Computing-Enabled IoT Networks
Aidaros, O. A., Kardjadja, Y., Bouida, Z., & Ibnkahla, M. A. (2023) · IEEE Sensors Applications Symposium (SAS) · doi:10.1109/sas58821.2023.10254051 · Cited by 3
Impact of Platt Scaling on Calibration in ML-Based Wireless Resource Allocation
Raina, R. K., Simmons, N., Simmons, D. E., Yacoub, M. D., & Cotton, S. L. (2025) · IEEE International Conference on Machine Learning for Communication and Networking (ICMLCN) · doi:10.1109/icmlcn64995.2025.11140554 · Cited by 3
Deep MIMO Detection Scheme for High-Speed Railways with Wireless Big Data
Chen, Z., Li, D., & Xu, Y. (2019) · IEEE Vehicular Technology Conference · doi:10.1109/vtcspring.2019.8746373 · Cited by 3
“In particular, it has been successfully applied in localization based on channel state information(CSI) [13], channel equalization [14], channel decoding [15] and resource allocation [16].”
MethodZhang, Y., Xia, W., Yan, F., Cheng, H., & Shen, L. (2019) · Lecture notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering · doi:10.1007/978-3-030-37262-0_12 · Cited by 4
“In [14], an intelligent agent is designed to develop a real-time adaptive policy for computational resource allocation in order to improve the average end-to-end reliability by employing a deep reinforcement learning method.”
BackgroundA Learning-Based Bandwidth Resource Allocation Method in Sliced 5G C-RAN
Sun, Y., Wang, Y., Yu, H., Guo, B., & Zhang, X. (2019) · IEEE Globecom Workshops (GC Wkshps) · doi:10.1109/gcwkshps45667.2019.9024681 · Cited by 3
“Besides, Reinforcement Learning (RL) is widely adopted in solving resource allocation problem, both [13] and [14] use the RL based methods to get the resource allocation problem settled.”
MethodHaberleşme Sistemlerinde Derin Öğrenme
Yıldırım, M. (2020) · European Journal of Science and Technology · doi:10.31590/ejosat.679929 · Cited by 3
Performance Analysis of Classifier Techniques at the Edge Node
Chandak, A. V., Ray, N. K., & Puthal, D. (2021) · IEEE International Symposium on Smart Electronic Systems (iSES) · doi:10.1109/ises52644.2021.00103 · Cited by 3
Survey of Edge Computing Based on a Generalized Framework and Some Recommendation
Sun, Y., Zhang, B., & Luo, M. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-59824-2_10 · Cited by 2
Li, Q., Cheng, H. B., Yang, Y., Tang, H., Liu, Z., Cao, Y., & Sun, W. (2022) · Lecture notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering · doi:10.1007/978-3-030-94763-7_4 · Cited by 2
Meng, L. (2022) · Journal of Sensors · doi:10.1155/2022/4418231 · Cited by 2
Wenjiao, W., Rongzuo, G., & Xiangkui, F. (2023) · IEEE Symposium on Computers and Communications (ISCC) · doi:10.1109/iscc58397.2023.10217987 · Cited by 2
“In order to validate the effectiveness of the improvedDDPG algorithm, we modified the AC algorithm, the DQNbased solution proposed in [21], and the DDPG-based solution proposed in [22] to make them applicable to our model, and used them as the comparative algorithms for the improved-DDPG.”
MethodTokuda, K., Sato, T., & Oki, E. (2023) · IEICE Transactions on Communications · doi:10.1587/transcom.2023ebp3043 · Cited by 2
Chen, B., Wu, G., Zhang, Q., & Tao, X. (2020) · Advances in intelligent systems and computing · doi:10.1007/978-981-15-8462-6_160 · Cited by 2
Huo, Y., Song, C., Ji, X., Yang, M., Yu, P., Tao, M., & Shi, L. (2020) · IEEE International Symposium on Broadband Multimedia Systems and Broadcasting (BMSB) · doi:10.1109/bmsb49480.2020.9379443 · Cited by 2
“How to link basic elements with optimization goals is very important[4-5].”
BackgroundTowards Efficient Edge Computing Through Adoption of Reinforcement Learning Strategies: A Review
Ray, A., & Chakrabarti, A. (2022) · Lecture notes on data engineering and communications technologies · doi:10.1007/978-981-19-2600-6_17 · Cited by 1
To Trust or Not to Trust: On Calibration in ML-Based Resource Allocation for Wireless Networks
Raina, R., Simmons, N., Simmons, D. E., Yacoub, M. D., & Duong, T. Q. (2025) · IEEE Transactions on Network Science and Engineering · doi:10.1109/tnse.2025.3636073 · Cited by 4
“Additionally, several studies have proposed ML-driven strategies for resource allocation in URLLC systems [38], [39], as well as for managing the coexistence of URLLC and enhanced mobile broadband services in 5G New Radio networks [40], [41], [42].”
Pettorali, M., Righetti, F., Vallati, C., Das, S. K., & Anastasi, G. (2026) · IEEE Internet of Things Journal · doi:10.1109/jiot.2026.3687394 · Cited by 1
“In this article, we focus on online approaches [8], [9], [10], [11], [12], [13], [14], [15], [16], which can dynamically manage system reconfigurations, and can be categorized as follows based on their methodology.”
Dealer: An Efficient Pricing Strategy for Deep-Learning-as-a-Service
Borjigin, W., Ota, K., & Dong, M. (2019) · International Conference on Communications · doi:10.1109/icc.2019.8761086 · Cited by 1
Artificial Intelligence for Optimizing Edge
Wang, X., Han, Y., Leung, V. C. M., Niyato, D. T., Yan, X., & Chen, X. (2020) · Springer eBooks · doi:10.1007/978-981-15-6186-3_8 · Cited by 1
Balanced Cloud Edge Resource Allocation Based on Conflict Conditions
Zhu, L., Feng, J., Liu, D., Yang, H., Bai, Z., Song, X., & Li, L. (2020) · IEEE Access · doi:10.1109/access.2020.3032428 · Cited by 1
“Baseon the mixed integer nonlinear programming problem of the approximate optimal protest model of resource allocation strategy was proposed [4], the strategy takes the minimum value of each mobile user’s transmission rate as the condition, establishes a fair resource allocation strategy with the…”
Evolving the Edge and the Cloud: A hybrid computing paradigm
Wang, J. (2021) · IEEE International Conference on Software Quality, Reliability and Security Companion (QRS-C) · doi:10.1109/qrs-c55045.2021.00108 · Cited by 1
Stateless ARE: Action Recommendation Engine without Network State Measurement
Mohammed, A. R., Mohammed, S., Côté, D., & Shirmohammadi, S. (2022) · IEEE International Symposium on Measurements & Networking (M&N) · doi:10.1109/mn55117.2022.9887710 · Cited by 1
“For example, in the complex Edge-User Allocation problem in cloud networking, RL [15] and Deep RL (DRL) [16] have been shown to perform very well in dynamic networking systems.”
BackgroundQ Learning Algorithm for Network Resource Management in Vehicular Communication Network
Agarwal, V., & Sharma, S. (2022) · Wiley eBooks · doi:10.1002/9781394152636.ch13 · Cited by 2
Aghasi, A., & Rituraj, R. (2022) · IEEE International Conference and Workshop Óbuda on Electrical and Power Engineering (CANDO-EPE) · doi:10.1109/cando-epe57516.2022.10046357 · Cited by 1
“In [13] authors use DRL to allocate computation resources of edge server to offloaded task.”
MethodA Deep Reinforcement Learning based policy gradient for Energy Consumption in Edge Computing
Saranya, G., & Sasikala, E. (2022) · International Conference on Computer, Power and Communications (ICCPC) · doi:10.1109/iccpc55978.2022.10072146 · Cited by 1
“In many research areas the DRL provides solutions for computational offloading and resource allocation [7].”
BackgroundLee, C., Lee, K., Cho, G., & Youn, C. (2024) · Communications in computer and information science · doi:10.1007/978-3-031-50385-6_5 · Cited by 1
PBRAMEC: Prioritized Buffer Based Resource Allocation for Mobile Edge Computing Devices
Hegde, S., Shravan, M. T., & Mukhopadhyay, A. (2024) · Lecture notes in networks and systems · doi:10.1007/978-981-99-6346-1_15 · Cited by 1
A dueling DQN-based online resource allocation algorithm for cloud computing
Zang, W., Li, S., & Yang, X. (2023) · International Conference on Communication Networks and Machine Learning (CNML) · doi:10.1145/3640912.3640986 · Cited by 1
Performance Enhancement and Scheduling in Communication Networks—A Review into Various Approaches
Kumari, P., & Jain, N. K. (2024) · Lecture notes in networks and systems · doi:10.1007/978-981-99-9562-2_55 · Cited by 2
Automated Saturation Mitigation Controlled by Deep Reinforcement Learning
Aguas, E., Lambert, A., Blanc, G., & Debar, H. (2020) · IEEE International Conference on Network Protocols (ICNP) · doi:10.1109/icnp49622.2020.9259356
Adaptive Collaborative Computing in Edge Computing Environment
Ren, J., Wang, H., & Zhang, X. (2020) · Lecture notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering · doi:10.1007/978-3-030-63941-9_12
Optimization of Workload Balancing and Power Allocation for Wireless Distributed Computing
Sun, C. K., Gao, X., & Ding, Z. J. (2022) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2022.3160493 · Cited by 1
“In [21], the authors investigated strategies for allocating computational resources using deep reinforcement learning in mobile edge computing networks.”
BackgroundWang, J., & Wang, J. (2022) · International Conference on Information Science and Technology · doi:10.1109/icist55546.2022.9926802
“Machine learning methods such as deep reinforcement learning approaches [12] are suitable for resource assignment.”
MethodLi, J., Yan, J., Yin, B., Zhao, L., & Wei, R. (2022) · IEEE International Conference on Internet of Things (iThings) and IEEE Green Computing and Communications (GreenCom) and IEEE Cyber, Physical and Social Computing (CPSCom) and IEEE Smart Data (SmartData) · doi:10.1109/ithings-greencom-cpscom-smartdata-cybermatics55523.2022.00115
“The allocation strategy of computation resources in the edge computation network has been proposed in [9] that solved the end-to-end reliability problem by the error probability.”
BackgroundHu, H., Wang, J., Wang, J., Wang, S., Wang, Y., Ye, Y., & Liao, R. (2022) · IEEE International Conference on Internet of Things (iThings) and IEEE Green Computing and Communications (GreenCom) and IEEE Cyber, Physical and Social Computing (CPSCom) and IEEE Smart Data (SmartData) · doi:10.1109/ithings-greencom-cpscom-smartdata-cybermatics55523.2022.00113
“In addition, learning based designs [17], [18] have been provided to”
Deep Reinforcement Learning for Downlink Resource Allocation in Vehicular Small Cell Networks
Brahmi, I., Hamdi, M., Rahmany, I., & Zarai, F. (2022) · IEEE International Symposium on Network Computing and Applications (NCA) · doi:10.1109/nca57778.2022.10013505
“Indeed, the function Q is approximated by a DNN with weights θ as a network Q [11].”
BackgroundMethodTwo-layer computation offloading strategy under collaborative D2D-assisted MEC systemPreprint
Xu, F., & Ning, L. (2023) · Research Square · doi:10.21203/rs.3.rs-2483291/v1
Predator-Prey Model Based Resource Allocation in Edge ComputingPreprint
liu, B., Bai, Y., & Ding, L. (2023) · Research Square · doi:10.21203/rs.3.rs-1436493/v1
Age-Based Metrics for Joint Control and Communication in Cyber-Physical Industrial Systems
Ana, P. M. D. S. (2023) · VBN Forskningsportal (Aalborg Universitet) · doi:10.54337/aau548863844
Optimization Research on Resource Allocation Strategy for User Mobility Perception in MecPreprint
Tian, P., Si, G., An, Z., Li, J., Liang, X., & Zhou, F. (2023) · SSRN Electronic Journal · doi:10.2139/ssrn.4576662
Adaptive Joint IoT device/Edge Computation Offloading for Heterogeneous IoT Applications
Aidaros, O. A. (2023) · Carleton University's Institutional Repository (MacOdrum Library, Carleton University) · doi:10.22215/etd/2023-15679
Yıldız, Ö., & SOKULLU, R. İ. (2023) · Telecommunications Forum (TELFOR) · doi:10.1109/telfor59449.2023.10372821
“For downlink concerns, [1] introduces a Deep Reinforcement Learning (DRL) algorithm to ensure task success while adhering to edge-to-edge delay and decoding error constraints.”
MethodAn Adaptive Type of Resource Allocation in the IOT Systems Using Reinforcement Type of Approach
Reddy, B. (2023) · International Conference on Power Energy, Environment and Intelligent Control (PEEIC) · doi:10.1109/peeic59336.2023.10450248
Wang, J., Chen, J., Zhou, Z. C., Zhou, D., Li, M., Yu, B., & Guan, Y. (2023) · IEEE International Conference on Internet of Things (iThings) and IEEE Green Computing and Communications (GreenCom) and IEEE Cyber, Physical and Social Computing (CPSCom) and IEEE Smart Data (SmartData) · doi:10.1109/ithings-greencom-cpscom-smartdata-cybermatics60724.2023.00073
“Therefore, mobile edge computing (MEC) [3], [4] has emerged as a promising technology to tackle this challenge by providing additional computation resources, which is instrumental in reducing latency and conversing bandwidth towards the backbone.”
BackgroundRegression-based investigation of latency in automated networks
Khan, U., Soomro, T. R., & Brohi, M. N. (2025) · Mehran University Research Journal of Engineering and Technology · doi:10.22581/muet1982.0339
Joint offloading-scheduling policies for future generation wireless networks
Djemai, I. (2024) · theses.fr (ABES) · doi:10.70675/69ec5992z792ez42bbz804cz4e953edbb5a8
Information theory and reinforcement learning of mixed covert and non-covert wireless networks
Bounhar, A. (2024) · theses.fr (ABES) · doi:10.70675/f9afa67bzfc94z4c01z9992za4da7b6e5bd5
Prédiction des besoins pour la gestion de serveurs mobiles en périphérie
Santi, N. (2023) · theses.fr (ABES) · doi:10.70675/9c8131fbze5abz4134z9452zf7c2f3c6e0da
A Survey of Cloud-Edge Collaborative Scheduling Strategies for Cyber-Physical Production Systems
Yu, X., Xia, C., Zeng, J., Xu, C., & Jin, X. H. (2026) · ACM Transactions on Embedded Computing Systems · doi:10.1145/3820365
“The energy consumption under cloud-edge collaboration includes local energy consumption and transmission energy consumption.”
Metadata
- Authors
- Yang, T., Hu, Y., Gursoy, M. C., Schmeink, A., & Mathar, R.
- Affiliations
- RWTH Aachen University (DE)
- Syracuse University (US)
- Venue
- 2018 15th International Symposium on Wireless Communication Systems (ISWCS)
- Publisher
- IEEE
- Published
- 2018-08
- Pages
- 1-5
- OpenAlex
- W2896079731
- Cited by
- 156
- References
- 15
- Citations by year
- 2026: 1 · 2025: 6 · 2024: 15 · 2023: 26 · 2022: 33 · 2021: 25 · 2020: 34 · 2019: 16
- FWCI
- 15.50
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- International Symposium on Wireless Communication Systems (publishedVersion)
- RWTH Publications (RWTH Aachen) (submittedVersion)
- Topics
- Physical Sciences › Computer Science › Computer Networks and Communications › IoT and Edge/Fog Computing
- Physical Sciences › Engineering › Electrical and Electronic Engineering › Advanced Wireless Network Optimization
- Physical Sciences › Engineering › Aerospace Engineering › Satellite Communication Systems
- Keywords
- resource allocation, mobile edge computing, deep Q-learning, finite blocklength codes, low-latency communication, task offloading, end-to-end reliability, delay violation probability, decoding error probability
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.