Amazing technological breakthrough possible @S-Logix

Office Address

  • 2nd Floor, #7a, High School Road, Secretariat Colony Ambattur, Chennai-600053 (Landmark: SRM School) Tamil Nadu, India
  • +91- 81240 01111

Social List

Deep Learning with Long Short-Term Memory for IoT Traffic Prediction - 2020

Deep Learning With Long Short-Term Memory For Iot Traffic Prediction

Research Area:  Machine Learning


5G network is new wireless mobile communication technology beyond 4G networks. These days, many network applications have been emerged and have led to an enormous amount of network traffic. Numerous studies have been conducted for enhancing the prediction accuracy of network traffic applications. Network traffic management and monitoring require technology for traffic prediction without the need for network operators. It is expected that each of the 5G networks and the Internet of things technologies to spread widely in the next few years. On the practical level, 5G uses the Internet of Things (IoT) for working in high-traffic networks with multiple sensors sending their packets to a destination simultaneously, which is an advantage of IoT applications. 5G presents wide bandwidth, low delay, and extremely high data throughput. Predicting network traffic has a great influence on IoT networks which results in reliable communication. A fully functional 5G network will not occur without artificial intelligence (AI) that can learn and make decisions on its own. Deep learning has been successfully applied to traffic prediction where it promotes traffic predictions via powerful fair representation learning. In this paper, we perform the prediction of IoT traffic in time series using LSTM - deep learning. the prediction accuracy has been evaluated using the RMSE as a merit function and mean absolute percentage error (MAPE).


Author(s) Name:  Ali R. Abdellah,Andrey Koucheryavy

Journal name:  

Conferrence name:  Internet of Things, Smart Spaces, and Next Generation Networks and Systems

Publisher name:  Springer

DOI:  10.1007/978-3-030-65726-0_24

Volume Information:  pp 267-280