WCSE 2015
ISBN: 978-981-09-5471-0 DOI: 10.18178/wcse.2015.04.109

Network Traffic Prediction based on Adaptive Neural Fuzzy Inference Systems

Yunjian Jia, Beili Wan, Liang Tang

Abstract— Due to the complexity of network traffic caused by the diversity of mobile applications and services, conventional traffic prediction methods face a great challenge to catch the characteristics of mobile networks. In this paper, we propose a network traffic prediction scheme based on adaptive neural fuzzy inference systems (ANFIS). Using actual data in mobile networks, we perform a training to adaptively adjust the parameters of fuzzy inference systems in order to get a model which gives accurate description of data characteristics. With the model, the mobile network traffic is predicted. The prediction performance is evaluated by computer simulation. Numerical results show that the root mean square error (RMSE) of training model is 0.0314 and the RMSE of traffic prediction is 0.0233. The good accuracy of the network traffic prediction may support future mobile network planning and resource allocation.

Index Terms— mobile networks, traffic prediction, adaptive neural fuzzy inference systems, actual data.

Yunjian Jia, Beili Wan, Liang Tang
College of Communication Engineering, Chongqing University, CHINA

[Download]


Cite:Yunjian Jia, Beili Wan, Liang Tang, "Network Traffic Prediction based on Adaptive Neural Fuzzy Inference Systems, " 2015 The 5th International Workshop on Computer Science and Engineering-Information Processing and Control Engineering (WCSE 2015-IPCE), pp. 675-681, Moscow, Russia, April 15-17, 2015.