Traffic forecasting in cellular networks using the LSTM RNN

In this work we design and implement a Neural Network that can identify recurrent patterns in various metrics which can be then used for cellular network traffic forecasting. Based on a custom architecture and memory, this Neural Network can handle prediction tasks faster and more accurately in real life scenarios. This approach offers a solution for service providers to enhance cellular network performance, by utilizing effectively the available resources. In order to provide a robust conclusion about the performance and precision of the proposed Neural Network, multiple predictions were made using the same data-set and the results were compared against other similar algorithms from the literature.

[1]  A. Neeraja,et al.  Licensed under Creative Commons Attribution Cc by Improving Network Management with Software Defined Networking , 2022 .

[2]  Bhaskar Krishnamachari,et al.  Energy Savings through Dynamic Base Station Switching in Cellular Wireless Access Networks , 2010, 2010 IEEE Global Telecommunications Conference GLOBECOM 2010.

[3]  Emile Fiesler,et al.  The Interchangeability of Learning Rate and Gain in Backpropagation Neural Networks , 1996, Neural Computation.

[4]  Zhisheng Niu,et al.  TANGO: traffic-aware network planning and green operation , 2011, IEEE Wireless Communications.

[5]  Albert-László Barabási,et al.  Limits of Predictability in Human Mobility , 2010, Science.

[6]  Nick Feamster,et al.  Improving network management with software defined networking , 2013, IEEE Commun. Mag..

[7]  Jian Liu.  Fractal Network Traffic Analysis with Applications , 2006 .

[8]  Samir Ranjan Das,et al.  Understanding traffic dynamics in cellular data networks , 2011, 2011 Proceedings IEEE INFOCOM.

[9]  Zhifeng Zhao,et al.  The predictability of cellular networks traffic , 2012, 2012 International Symposium on Communications and Information Technologies (ISCIT).

[10]  Zhisheng Niu,et al.  Toward dynamic energy-efficient operation of cellular network infrastructure , 2011, IEEE Communications Magazine.

[11]  A. Adas,et al.  Traffic models in broadband networks , 1997, IEEE Commun. Mag..

[12]  Jordi Domingo-Pascual,et al.  Evaluation of network traffic prediction based on neural networks with multi-task learning and multiresolution decomposition , 2011, 2011 IEEE 7th International Conference on Intelligent Computer Communication and Processing.

[13]  A. Lapedes,et al.  Nonlinear Signal Processing Using Neural Networks , 1987 .

[14]  Manish Joshi,et al.  A Review of Network Traffic Analysis and Prediction Techniques , 2015, ArXiv.

[15]  Evgenia F. Adamopoulou,et al.  On the predictability of next generation mobile network traffic using artificial neural networks , 2015, Int. J. Commun. Syst..

[16]  A. Lapedes,et al.  Nonlinear signal processing using neural networks: Prediction and system modelling , 1987 .

[17]  Honggang Zhang,et al.  Spatial modeling of the traffic density in cellular networks , 2014, IEEE Wireless Communications.

[18]  Meng Qing-Fang,et al.  Small-time scale network traffic prediction based on a local support vector machine regression model , 2009 .

[19]  Bo Zhou,et al.  Network Traffic Modeling and Prediction with ARIMA / GARCH , 2005 .

[20]  A. Wolisz,et al.  Primary Users in Cellular Networks: A Large-Scale Measurement Study , 2008, 2008 3rd IEEE Symposium on New Frontiers in Dynamic Spectrum Access Networks.

[21]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..