Real-time Processing Optimization of Convolutional Neural Network in Edge Computing System
Keywords:
Edge computing, Convolutional neural networks, Real-time processing, Network traffic analysis, Apache KafkaAbstract
On the edge nodes of Azure Stack Edge, network interfaces often experience packet queuing during peak periods due to congestion and high load. To address this issue, MobileNet was used to analyze network traffic in real-time and detect congestion, aiming to balance node network load and reduce latency. Firstly, data was collected and key features were filtered during peak periods using Wireshark, including packet rate, bidirectional packet count, flag statistics, etc.; next, TensorFlow Lite was used to prune and lightweight the model, removing redundant network layers and reducing floating-point operation accuracy, making it suitable for edge devices; then, based on the optimized model, real-time classification of network traffic was performed using Apache Kafka to detect potential congestion situations; at the same time, the load balancing strategy was dynamically adjusted based on the detection results to optimize queue priority and traffic transfer, thereby alleviating congestion. The results showed that during peak hours, the average processing time for ICMP (Internet Control Message Protocol), TCP (Transmission Control Protocol), and UDP (User Datagram Protocol) traffic was reduced by 27.1%, 23.1%, and 20.7%, respectively. When the cache capacity was 16GB, the hit rate reached 94%. Under standard traffic conditions during peak hours, the packet loss rate dropped to 2.1%, which could significantly improve the performance of edge node networks.
References
[1]Liu GX, Shi H, Kiani A, Khreishah A, Lee J, et al. Smart traffic monitoring system using computer vision and edge computing. IEEE Transactions on Intelligent Transportation Systems. 2022, 23(8), 12027-12038. DOI: 10.1109/TITS.2021.3109481
[2]Carvalho G, Cabral B, Pereira V, Bernardino J. Edge computing: current trends, research challenges and future directions. Computing. 2021, 103(5), 993-1023. DOI: 10.1007/s00607-020-00896-5
[3]Yuan XM, Chen JH, Yang JY, Zhang N, Yang TT, et al. Fedstn: Graph representation driven federated learning for edge computing enabled urban traffic flow prediction. IEEE Transactions on Intelligent Transportation Systems. 2023, 24(8), 8738-8748. DOI: 10.1109/TITS.2022.3157056
[4]Wang Z, Hu J, Min GY, Zhao ZW, Wang J. Data-augmentation-based cellular traffic prediction in edge-computing-enabled smart city. IEEE Transactions on Industrial Informatics. 2021, 17(6), 4179-4187. DOI: 10.1109/TII.2020.3009159
[5]Bi J, Zhang X, Yuan HT, Zhang J, Zhou MC. A hybrid prediction method for realistic network traffic with temporal convolutional network and LSTM. IEEE Transactions on Automation Science and Engineering. 2022, 19(3), 1869-1879. DOI: 10.1109/TASE.2021.3077537
[6]Zhang YN, Zhou YH, Lu HP, Fujita H. Traffic network flow prediction using parallel training for deep convolutional neural networks on spark cloud. IEEE Transactions on Industrial Informatics. 2020, 16(12), 7369-7380. DOI: 10.1109/TII.2020.2976053
[7]Yu A, Yang H, Nguyen KK, Zhang J, Cheriet M. Burst traffic scheduling for hybrid E/O switching DCN: An error feedback spiking neural network approach. IEEE Transactions on Network and Service Management. 2021, 18(1), 882-893. DOI: 10.1109/TNSM.2020.3040907
[8]Hu L, Miao YM, Yang J, Ghoneim A, Hossain MS, et al. If-rans: intelligent traffic prediction and cognitive caching toward fog-computing-based radio access networks. IEEE Wireless Communications. 2020, 27(2), 29-35. DOI: 10.1109/MWC.001.1900368
[9]Serdaroglu KC, Baydere S. An efficient multipriority data packet traffic scheduling approach for fog of things. IEEE Internet of Things Journal. 2022, 9(1), 525-534. DOI: 10.1109/JIOT.2021.3084502
[10]Admassu T, Bizuneh HD. Improving network performance with an integrated priority queue and weighted fair queue scheduling. Indonesian Journal of Electrical Engineering and Computer Science. 2020, 19(1), 241-247. DOI: 10.11591/ijeecs.v19.i1.pp241-247
[11]Samal T, Kabat MR. A prioritized traffic scheduling with load balancing in wireless body area networks. Journal of King Saud University-Computer and Information Sciences. 2022, 34(8), 5448-5455. DOI: 10.1016/j.jksuci.2020.12.023
[12]Yang Z, Liu YW, Chen Y, Zhou JT. Deep learning for latent events forecasting in content caching networks. IEEE Transactions on Wireless Communications. 2022, 21(1), 413-428. DOI: 10.1109/TWC.2021.3096747
[13]Tang B, Kang LY. EICache: A learning-based intelligent caching strategy in mobile edge computing. Peer-to-Peer Networking and Applications. 2022, 15(2), 934-949. DOI: 10.1007/s12083-021-01266-4
[14]Hardegen C, Pfülb B, Rieger S, Gepperth A. Predicting network flow characteristics using deep learning and real-world network traffic. IEEE Transactions on Network and Service Management. 2020, 17(4), 2662-2676. DOI: 10.1109/TNSM.2020.3025131
[15]Wong ML, Arjunan T. Real-Time Detection of Network Traffic Anomalies in Big Data Environments Using Deep Learning Models. Emerging Trends in Machine Intelligence and Big Data. 2024, 16(2), 1-11.
[16]Wei GL, Wang ZH. Adoption and realization of deep learning in network traffic anomaly detection device design. Soft Computing. 2021, 25(2), 1147-1158. DOI: 10.1007/s00500-020-05210-1
[17]Singh K, Mahajan A, Mansotra V. 1D-CNN based Model for Classification and Analysis of Network Attacks. International Journal of Advanced Computer Science and Applications. 2021, 12(11), 604-613. DOI: 10.14569/IJACSA.2021.0121169
[18]Anitha T, Aanjankumar S, Poonkuntran S, Nayyar A. A novel methodology for malicious traffic detection in smart devices using BI-LSTM–CNN-dependent deep learning methodology. Neural Computing and Applications. 2023, 35(27), 20319-20338. DOI: 10.1007/s00521-023-08818-0
[19]Kumar R, Swarnkar M, Singal G, Kumar N. IoT network traffic classification using machine learning algorithms: An experimental analysis. IEEE Internet of Things Journal. 2022, 9(2), 989-1008. DOI: 10.1109/JIOT.2021.3121517
[20]Ullah F, Ullah S, Srivastava G, Lin JCW. IDS-INT: Intrusion detection system using transformer-based transfer learning for imbalanced network traffic. Digital Communications and Networks, 2024, 10(1), 190-204. DOI: 10.1016/j.dcan.2023.03.008
[21]Shahraki A, Abbasi M, Taherkordi A, Jurcut AD. Active learning for network traffic classification: a technical study. IEEE Transactions on Cognitive Communications and Networking. 2021, 8(1), 422-439. DOI: 10.1109/TCCN.2021.3119062
[22]Li WZ, Gao SH, Li X, Xu YT, Lu SL. Tcp-neuroc: Neural adaptive tcp congestion control with online changepoint detection. IEEE Journal on Selected Areas in Communications. 2021, 39(8), 2461-2475. DOI: 10.1109/JSAC.2021.3087247
[23]Salman O, Elhajj IH, Kayssi A, Chehab A. A review on machine learning–based approaches for Internet traffic classification. Annals of Telecommunications. 2020, 75(11), 673-710. DOI: 10.1007/s12243-020-00770-7
[24]Nascita A, Montieri A, Aceto G, Ciuonzo D, Persico V, et al. XAI meets mobile traffic classification: Understanding and improving multimodal deep learning architectures. IEEE Transactions on Network and Service Management. 2021, 18(4), 4225-4246. DOI: 10.1109/TNSM.2021.3098157
[25]Wang K, Yu XY, Lin WL, Deng ZL, Liu X. Computing aware scheduling in mobile edge computing system. Wireless Networks. 2021, 27(6), 4229-4245. DOI: 10.1007/s11276-018-1892-z
[26]Yang JY, Chen YX, Xue KP, Han JP, Li J, et al. IEACC: an intelligent edge-aided congestion control scheme for named data networking with deep reinforcement learning. IEEE Transactions on Network and Service Management. 2022, 19(4), 4932-4947. DOI: 10.1109/TNSM.2022.3196344
[27]Shingai R, Hiraga Y, Fukuoka H, Mitani T, Nakada T, Nakashima Y. Construction of an Efficient Divided/Distributed Neural Network Model Using Edge Computing. IEICE TRANSACTIONS on Information and Systems. 2020, 103(10), 2072-2082. DOI: 10.1587/transinf.2019EDP7326
[28]Zhao Y, Yin Y, Gui G. Lightweight deep learning based intelligent edge surveillance techniques. IEEE Transactions on Cognitive Communications and Networking. 2020, 6(4), 1146-1154. DOI: 10.1109/TCCN.2020.2999479
[29]Burra LR, Karuna A, Tumma S, Marlapalli K, Tumuluru P. MobileNetV2-based Transfer Learning Model with Edge Computing for Automatic Fabric Defect Detection. Journal of Scientific & Industrial Research. 2023, 82(1), 128-134. DOI: 10.56042/jsir.v82i1.69928
[30]Le KH, Le-Minh KH, Thai HT. Brainyedge: An ai-enabled framework for iot edge computing. ICT Express. 2023, 9(2), 211-221. DOI: 10.1016/j.icte.2021.12.007
[31]Cheng SB, Jin YF, Harrison SP, Quilodrán-Casas C, Prentice IC, et al. Parameter flexible wildfire prediction using machine learning techniques: Forward and inverse modelling. Remote Sensing. 2022, 14(13), 3228. DOI: 10.3390/rs14133228
[32]Zhuang YL, Cheng SB, Kovalchuk N, Simmons M, Matar OK, et al. Ensemble latent assimilation with deep learning surrogate model: application to drop interaction in a microfluidics device. Lab on a Chip. 2022, 22(17), 3187-3202. DOI: 10.1039/D2LC00303A
[33]Chen DY, Cheng SB, Hu JW, Kasoar M, Arcucci R. Explainable Global Wildfire Prediction Models using Graph Neural Networks. arXiv preprint arXiv:2402.07152, 2024. DOI: 10.48550/arXiv.2402.07152
[34]Laghari AA, Zhang XB, Shaikh ZA, Khan A, Estrela VV, et al. A review on quality of experience (QoE) in cloud computing. Journal of Reliable Intelligent Environments. 2024, 10(2), 107-121. DOI: 10.1007/s40860-023-00210-y
[35]Yamazaki T. Quality of experience (QoE) studies: Present state and future prospect. IEICE Transactions on Communications. 2021, 104(7), 716-724. DOI: 10.1587/transcom.2020CQI0003
[36]Agarwal B, Togou MA, Ruffini M, Muntean GM. Qoe-driven optimization in 5g o-ran-enabled hetnets for enhanced video service quality. IEEE Communications Magazine. 2023, 61(1), 56-62. DOI: 10.1109/MCOM.003.2200229
[37]Gelenbe E, Domanska J, Fröhlich P, Nowak MP, Nowak S. Self-aware networks that optimize security, QoS, and energy. Proceedings of the IEEE. 2020, 108(7), 1150-1167. DOI: 10.1109/JPROC.2020.2992559
[38]Putra FPE, Ubaidi U, Aziz M, Irfan M, Alim R. Improving Network Service Quality in parts of Sampang City: QoS Evaluation and User Perception of QoE. Brilliance: Research of Artificial Intelligence. 2024, 4(1), 408-412. DOI: 10.47709/brilliance.v4i1.4311
[39]Yuan YZ, Cao X, Liu ZX, Chen CL, Guan XP. Adaptive priority adjustment scheduling approach with response-time analysis in time-sensitive networks. IEEE Transactions on Industrial Informatics. 2022, 18(12), 8714-8723. DOI: 10.1109/TII.2022.3150044
[40]Velayutham A. Methods and algorithms for optimizing network traffic in next-generation networks: Strategies for 5g, 6g, sdn, and iot systems. Journal of Intelligent Connectivity and Emerging Technologies. 2021, 6(5), 1-26.
[41]Jung S, Kim J, Kim JH. Intelligent active queue management for stabilized QoS guarantees in 5G mobile networks. IEEE Systems Journal. 2020, 15(3), 4293-4302. DOI: 10.1109/JSYST.2020.3014231
[42]Anwar MR, Wang S, Akram MF, Raza S, Mahmood S. 5G-enabled MEC: A distributed traffic steering for seamless service migration of internet of vehicles. IEEE Internet of Things Journal. 2022, 9(1), 648-661. DOI: 10.1109/JIOT.2021.3084912
[43]Perera S, Gupta V, Buckley W. Management of online server congestion using optimal demand throttling. European Journal of Operational Research. 2020, 285(1), 324-342. DOI: 10.1016/j.ejor.2020.02.008
Downloads
Published
Issue
Section
License
Copyright (c) 2024 Journal of Computational Systems and Applications
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
