tuzsutТруды учебных заведений связиProceedings of Telecommunication Universities1813-324X2712-8830СПбГУТ10.31854/1813-324X-2022-8-4-89-99tuzsut-420Research ArticleРЕЗУЛЬТАТЫ ИССЛЕДОВАНИЙ МОЛОДЫХ УЧЕНЫХRESEARCH RESULTS BY YOUNG SCIENTISTSМетоды машинного обучения для прогнозирования трафика в многоуровневой облачной архитектуре для сервисов автономных транспортных средствDeep Learning Approaches for Traffic Prediction Forecasting in Multi-Level Cloud Architecture for Autonomous Vehicle Serviceshttps://orcid.org/0000-0002-6267-4727Аль-СвейтиМ.AlsweityM.

Аль-Свейти Малик, аспирант кафедры сети связи и передачи данных

Санкт-Петербург, 193232

Malik Alsweity

St. Petersburg, 193232

aldonasmar@gmail.comСанкт-Петербургский государственный университет телекоммуникаций им. М.А. Бонч-БруевичаРоссияThe Bonch-Bruevich Saint-Petersburg State University of TelecommunicationsRussian Federation202210012023848999Copyright © Аль-Свейти М., 20232023Аль-Свейти М.Alsweity M.This work is licensed under a Creative Commons Attribution 4.0 License.https://tuzs.sut.ru/jour/article/view/420

Автономные транспортные средства (AV, аббр. от англ. Autonomous Vehicle) являются одним из наиболее важных новых вариантов использования и перспективной технологией для сетей пятого поколения (5G) и следующего поколения в многочисленных приложениях. В настоящее время использование AV экспоненциально растет во всем мире, благодаря быстрому росту осведомленности и применению методов искусственного интеллекта в различных областях. Прогнозирование потоков данных необходимо для AV, чтобы улучшить передачу данных и уменьшить задержки за счет более эффективного использования соответствующих возможностей, мониторинга, управления и контроля дорожной системы. В данной работе предлагается подход глубокого обучения с двунаправленной моделью с долгой краткосрочной памятью (BI-LSTM, аббр. от англ. Bidirectional Long-Short-Term Memory) для прогнозирования сетевого трафика AV с многоуровневыми сервисами облачных вычислений. С точки зрения точности прогнозирования проводится сравнение между BI-LSTM и однонаправленной моделью с долгой краткосрочной памятью (LSTM) в зависимости от количества используемых пакетов. Точность предсказания рассчитывается с помощью среднеквадратичной ошибки, средней абсолютной процентной ошибки, коэффициента детерминации (R2) и времени обработки. Результаты показывают, что точность прогнозирования с помощью BI-LSTM превосходит модель LSTM. Кроме того, точность прогнозирования с использованием размера обучающей партии (BatchSize) равной 8, превосходит конкурентов и обеспечивает высокую производительность.

Autonomous vehicle (AV) is one of the most new use cases and a technology for fifth-generation (5G) and next-generation mobile networks in numerous applications., the use of AVs has exponentially worldwide due to the rapidly growing awareness and use of artificial intelligence (AI) methods in various fields. Predicting data flows is essential for AVs to improve data transmission and decrease delays through more efficient use of appropriate capabilities, monitoring, management, and control of the traffic system. This paper proposes a deep learning approach (DL) with the bidirectional long-short-term memory model (BI-LSTM) for predicting the traffic rates of AVs with multi-cloud services. In terms of prediction accuracy, a comparison is conducted between the BI-LSTM and the unidirectional LSTM based on the number of batch sizes used. The prediction accuracy is computed using the root mean square error (RMSE), mean absolute percentage error (MAPE), coefficient of determination (R2), and processing time. The results show that the prediction accuracy with BI-LSTM outperforms the LSTM model. Besides, the prediction accuracy using 8 batch sizes outperforms the competitors and offers outstanding performance.

беспилотные автомобили5GMECMLDLпрогнозированиеBI-LSTMLSTMautonomous vehicles5GMECMLDLpredictionBI-LSTMLSTMReferencesAbdellah A.R., Alshahrani A., Muthanna A., Koucheryavy A. Performance Estimation in V2X Networks Using Deep Learning-Based M-Estimator Loss Functions in the Presence of Outliers // Symmetry. 2021. Vol. 13. Iss. 11. P. 2207. DOI:10.3390/sym13112207.Abdellah A.R., Alshahrani A., Muthanna A., Koucheryavy A. Performance Estimation in V2X Networks Using Deep Learning-Based M-Estimator Loss Functions in the Presence of Outliers. Symmetry. 2021;13(11);2207. DOI:10.3390/sym13112207Chochliouros I.P., Spiliopoulou A.S., Lazaridis P., Dardamanis A., Zaharis Z., Kostopoulos A. Dynamic Network Slicing: Challenges and Opportunities // Proceedings of the International Conference on Artificial Intelligence Applications and Innovations (AIAI 2020, Neos Marmaras, Greece, 5–7 June 2020). IFIP WG 12.5 International Workshops. IFIP Advances in Information and Communication Technology. Vol. 585. Cham: Springer, 2020. DOI:10.1007/978-3-030-49190-1_5Chochliouros I.P., Spiliopoulou A.S., Lazaridis P., Dardamanis A., Zaharis Z., Kostopoulos A. Dynamic Network Slicing: Challenges and Opportunities. Proceedings of the International Conference on Artificial Intelligence Applications and Innovations, AIAI 2020, 5–7 June 2020, Neos Marmaras, Greece. IFIP WG 12.5 International Workshops. IFIP Advances in Information and Communication Technology, vol.585. Cham: Springer; 2020. DOI:10.1007/978-3-030-49190-1_5Miglani A., Kumar N. Deep larning models for traffic flow prediction in autonomous vehicles: A review, solutions, and challenges // Vehicular Communications. 2019. Vol. 20. P. 100184. DOI:10.1016/j.vehcom.2019.100184Miglani A., Kumar N. Deep learning models for traffic flow prediction in autonomous vehicles: A review, solutions, and challenges. Vehicular Communications. 2019;20:100184. DOI:10.1016/j.vehcom.2019.100184Gillani M., Niaz H.A., Farooq M.U., Ullah A. Data collection protocols for VANETs: a survey // Complex & Intelligent Systems. 2022. Vol. 8. Iss. 3. PP. 2593‒2622. DOI:10.1007/s40747-021-00629-xGillani M., Niaz H.A., Farooq M.U., Ullah A. Data collection protocols for VANETs: a survey. Complex & Intelligent Systems. 2022;8(3):2593‒2622. DOI:10.1007/s40747-021-00629-xGrigorescu S, Trasnea B, Cocias T, Macesanu G. A survey of deep learning techniques for autonomous driving // Journal of Field Robotics. 2020. Vol. 37. Iss. 3. PP. 362–386. DOI:10.1002/rob.21918Grigorescu S, Trasnea B, Cocias T, Macesanu G. A survey of deep learning techniques for autonomous driving. Journal of Field Robotics. 2020:37(3):362–386. DOI:10.1002/rob.21918Pfülb B., Hardegen C., Gepperth A., Rieger S. A Study of Deep Learning for Network Traffic Data Forecasting // Proceedings of the 28th International Conference on Artificial Neural Networks and Machine Learning (ICANN 2019, Munich, Germany, 17–19 September 2019). Lecture Notes in Computer Science. Vol. 11730. Cham: Springer, 2019. PP. 497–512. DOI:10.1007/978-3-030-30490-4_40Pfülb B., Hardegen C., Gepperth A., Rieger S. A Study of Deep Learning for Network Traffic Data Forecasting. Proceedings of the 28th International Conference on Artificial Neural Networks and Machine Learning, ICANN 2019, 17–19 September 2019, Munich, Germany. Lecture Notes in Computer Science, vol.11730. Cham: Springer; 2019. p.497–512. DOI:10.1007/978-3-030-30490-4_40Ateeq M., Ishmanov F., Afzal M.K., Naeem M. Predicting Delay in IoT Using Deep Learning: A Multiparametric Approach // IEEE Access. 2019. Vol. 7. PP. 62022‒62032. DOI:10.1109/ACCESS.2019.2915958Ateeq M., Ishmanov F., Afzal M.K., Naeem M. Predicting Delay in IoT Using Deep Learning: A Multiparametric Approach. IEEE Access. 2019;7:62022‒62032. DOI:10.1109/ACCESS.2019.2915958Abdellah A.R., Koucheryavy A. Deep Learning with Long Short-Term Memory for IoT Traffic Prediction // Proceedings of the 20th International Conference on Next Generation Networks and Systems, NEW2AN 2020, and 13th Conference on Internet of Things, Smart Spaces, ruSMART 2020 (St. Petersburg, Russia, 26–28 August 2020. Lecture Notes in Computer Science. Vol. 12525. Cham: Springer, 2020. PP. 267–280. DOI:10.1007/978-3-030-65726-0_24Abdellah A.R., Koucheryavy A. Deep Learning with Long Short-Term Memory for IoT Traffic Prediction. Proceedings of the 20th International Conference on Next Generation Networks and Systems, NEW2AN 2020, and 13th Conference on Internet of Things, Smart Spaces, ruSMART 2020, 26–28 August 2020, St. Petersburg, Russia. Lecture Notes in Computer Science, vol.12525. Cham: Springer; 2020. p.267–280. DOI:10.1007/978-3-030-65726-0_24Almutairi M.S. Deep Learning-Based Solutions for 5G Network and 5G-Enabled Internet of Vehicles: Advances, Meta-Data Analysis, and Future Direction // Mathematical Problems in Engineering. 2022. Vol. 2022. P. 6855435. DOI:10.1155/2022/6855435Almutairi M.S. Deep Learning-Based Solutions for 5G Network and 5G-Enabled Internet of Vehicles: Advances, Meta-Data Analysis, and Future Direction. Mathematical Problems in Engineering. 2022;2022:6855435. DOI:10.1155/2022/6855435Kaur J., Khan M.A., Iftikhar M., Imran M., Haq Q.E.U. Machine Learning Techniques for 5G and Beyond // IEEE Access. 2021. Vol. 9. PP. 23472‒23488. DOI:10.1109/ACCESS.2021.3051557Kaur J., Khan M.A., Iftikhar M., Imran M., Haq Q.E.U. Machine Learning Techniques for 5G and Beyond. IEEE Access. 2021;9:23472‒23488. DOI:10.1109/ACCESS.2021.3051557Abdellah A.R., Volkov A., Muthanna A., Gallyamov D., Koucheryavy A. Deep Learning for IoT Traffic Prediction Based on Edge Computing // Proceedings of the 23rd International Conference on Distributed Computer and Communication Networks: Control, Computation, Communications (DCCN 2020, Moscow, Russia, 14‒18 September 2020). Com munications in Computer and Information Science. Vol. 1337. PP. 18‒29. Cham: Springer, 2020. DOI:10.1007/978-3-030-66242-4_2Abdellah A.R., Volkov A., Muthanna A., Gallyamov D., Koucheryavy A. Deep Learning for IoT Traffic Prediction Based on Edge Computing. Proceedings of the 23rd International Conference on Distributed Computer and Communication Networks: Control, Computation, Communications, DCCN 2020, 14‒18 September 2020, Moscow, Russia. Communications in Computer and Information Science, vol.1337. p.18‒29. Cham: Springer, 2020. DOI:10.1007/978-3-030-66242-4_2Khedkar S.P., Canessane R.A., Najafi M.L. Prediction of Traffic Generated by IoT Devices Using Statistical Learning Time Series Algorithms // Wireless Communications and Mobile Computing. 2021. Vol. 2021. P. 5366222. DOI:10.1155/2021/5366222Khedkar S.P., Canessane R.A., Najafi M.L. Prediction of Traffic Generated by IoT Devices Using Statistical Learning Time Series Algorithms. Wireless Communications and Mobile Computing. 2021;2021:5366222. DOI:10.1155/2021/5366222Selvamanju E., Shalini V.B. Machine Learning Based Mobile Data Traffic Prediction in 5G Cellular Networks // Proceedings of the 5th International Conference on Electronics, Communication and Aerospace Technology (ICECA, Coimbatore, India, 02‒04 December 2021). IEEE, 2021. PP. 1318‒1324. DOI:10.1109/ICECA52323.2021.9675887Selvamanju E., Shalini V.B. Machine Learning Based Mobile Data Traffic Prediction in 5G Cellular Networks. Proceedings of the 5th International Conference on Electronics, Communication and Aerospace Technology, ICECA, 02‒04 December 2021, Coimbatore, India. IEEE; 2021. p.1318‒1324. DOI:10.1109/ICECA52323.2021.9675887Choi D., Yim J., Baek M., Lee S. Machine Learning-Based Vehicle Trajectory Prediction Using V2V Communications and On-Board Sensors // Electronics. 2021. Vol. 10. Iss. 4. P. 420. DOI:10.3390/electronics10040420Choi D., Yim J., Baek M., Lee S. Machine Learning-Based Vehicle Trajectory Prediction Using V2V Communications and On-Board Sensors. Electronics. 2021;10(4):420. DOI:10.3390/electronics10040420Rasouli A., Tsotsos J.K. Autonomous Vehicles That Interact With Pedestrians: A Survey of Theory and Practice // IEEE Transactions on Intelligent Transportation Systems. 2020. Vol. 21. Iss. 3. PP. 900‒918. DOI:10.1109/TITS.2019.2901817Rasouli A., Tsotsos J.K. Autonomous Vehicles That Interact With Pedestrians: A Survey of Theory and Practice. IEEE Transactions on Intelligent Transportation Systems. 2020;21(3):900‒918. DOI:10.1109/TITS.2019.2901817Badicu A., Suciu G., Balanescu M., Dobrea M., Birdici A., Orza O., Pasat A. PMs concentration forecasting using ARIMA algorithm // Proceedings of the 91-st Vehicular Technology Conference (VTC2020-Spring, Antwerp, Belgium, 25–28 May 2020). IEEE, 2020. DOI:10.1109/VTC2020-Spring48590.2020.9129390Badicu A., Suciu G., Balanescu M., Dobrea M., Birdici A., Orza O., Pasat A. PMs concentration forecasting using ARIMA algorithm. Proceedings of the 91-st Vehicular Technology Conference, VTC2020-Spring, 25–28 May 2020, Antwerp, Belgium. IEEE; 2020. DOI:10.1109/VTC2020-Spring48590.2020.9129390Shang Q., Tan D., Gao S., Feng L. A Hybrid Method for Traffic Incident Duration Prediction Using BOA-Optimized Random Forest Combined with Neighborhood Components Analysis // Journal of Advanced Transportation. 2019. Vol. 2019. P. 4202735. DOI:10.1155/2019/4202735Shang Q., Tan D., Gao S., Feng L. A Hybrid Method for Traffic Incident Duration Prediction Using BOA-Optimized Random Forest Combined with Neighborhood Components Analysis. Journal of Advanced Transportation. 2019;2019:4202735. DOI:10.1155/2019/4202735Кучерявый А.Е. Сети связи с ультрамалыми задержками // Труды научно-исследовательского института радио. 2019. № 1. С. 69‒74.Kucheryavy A. Network Delay Communications Networks. Trudy NIIR. 2019;1:69‒74. (in Russ.)Мутханна А.С. Интеллектуальная распределенная архитектура сети связи для поддержки беспилотных автомобилей // Электросвязь. 2020. № 7. С. 29‒34. DOI:10.34832/ELSV.2020.8.7.004Muthanna A.S. Distributed Intelligent Communication Network Architecture for Unmanned Vehicles. Electrosvyaz. 2020;7:29‒34. (in Russ.) DOI: 10.34832/ELSV.2020.8.7.004Владыко А.Г., Мутханна А.С., Кучерявый А.Е. Метод выгрузки трафика в V2X/5G сетях на основе системы граничных вычислений // Электросвязь. 2020. № 8. С. 24‒30. DOI:10.34832/ELSV.2020.9.8.004Vladyko A.G., Muthanna A.S., Koucheryavy A.Eu. Traffic Offloading Method for V2X/5G Networks Based on the Edge Computing System. Electrosvyaz. 2020;8:24‒30. (in Russ.) DOI: 10.34832/ELSV.2020.9.8.004Shrestha R., Bajracharya R., Nam S.Y. Challenges of Future VANET and Cloud-Based Approaches // Wireless Communications and Mobile Computing. 2018. Vol. 2018. P. 5603518. DOI:10.1155/2018/5603518Shrestha R., Bajracharya R., Nam S.Y. Challenges of Future VANET and Cloud-Based Approaches. Wireless Communications and Mobile Computing. 2018;2018:5603518. DOI:10.1155/2018/5603518Плотников П.В., Владыко А.Г. Минимизация задержек при взаимодействии граничных устройств с использованием кластеризации в сетях VANETs // Труды учебных заведений связи. 2022. Т. 8. № 2. С. 6–13. DOI:10.31854/1813-324X-2022-8-2-6-13Plotnikov P., Vladyko A. Minimizing Delays in the Interaction of Edge Devices Using Clustering in VANETs. Proceedings of Telecommunication Universities. 2022;8(2):6‒13. (in Russ.) DOI:10.31854/1813-324X-2022-8-2-6-13

The authors declare that there are no conflicts of interest present.