Artificial Neural Network Non-linear Auto Regressive Moving Average (NARMA) Model for Internet Traffic Prediction

Authors

  • Mohd Naqiuddin Sahrani Faculty of Electrical Engineering, Universiti Teknologi Mara, Shah Alam, Malaysia
  • Md Mahfudz Mat Zan Faculty of Electrical Engineering, Universiti Teknologi Mara, Shah Alam, Malaysia
  • Ihsan Mohd Yassin Faculty of Electrical Engineering, Universiti Teknologi Mara, Shah Alam, Malaysia
  • Azlee Zabidi Faculty of Electrical Engineering, Universiti Teknologi Mara, Shah Alam, Malaysia
  • Megat Syahirul Amin Megat Ali Faculty of Electrical Engineering, Universiti Teknologi Mara, Shah Alam, Malaysia

Keywords:

Network Traffic Prediction, Nonlinear AutoRegressive Moving Average (NARMA), System Identification, Forecasting,

Abstract

The technology of computing and network communication is undergoing rapid development, leading to increasing number of applications and services being available online. As more applications are available online, network traffic becomes a significant problem as high network loads may limit access to users. In this paper, we propose an internet traffic Nonlinear Auto-Regressive Moving Average model (NARMA) prediction model to assist network managers in forecasting internet traffic and planning their resources accordingly. The Multi-Layer Perceptron (MLP) estimator was used in this paper. The performance of the model were evaluated using Mean Squared Error (MSE), correlation tests, and residual histogram tests with good agreement between the model and actual outputs.

Downloads

Published

2017-03-15

How to Cite

Sahrani, M. N., Mat Zan, M. M., Mohd Yassin, I., Zabidi, A., & Megat Ali, M. S. A. (2017). Artificial Neural Network Non-linear Auto Regressive Moving Average (NARMA) Model for Internet Traffic Prediction. Journal of Telecommunication, Electronic and Computer Engineering (JTEC), 9(1-3), 145–149. Retrieved from https://jtec.utem.edu.my/jtec/article/view/1760