Predictive and Comparative Analysis of NARX and NIO Time Series Prediction

Journal Title: American journal of Engineering Research - Year 2017, Vol 6, Issue 9

Abstract

Telecommunication is one of the most unavoidable utilities of daily life of human being. It provides data records of its services from call logs, port activities to internet usage or consumption. This paper investigates and explored the methodologies for modeling, simulation and controls in ANN based of time series application of telecommunication. To show and prove efficiency, simulated and operational data sets are employed to demonstrate the capability of neural networks in capturing complex nonlinear dynamics where NARX and NIO models are set up to explore and compare both steady-state and transient features on daily internet usage activities. The structures were configured, generated and run in MATLAB to create and train platform, validation, testing and results demonstrate that the techniques can be applied accurately which means that both models successfully capture dynamics of the system up to a certain degree of acceptance. The related parameters for the design and simulation are tuned and set up according to the requirements which show that ANN can perform even better than conventional methods. Finally, it was deduced that NARX model outperform more than the NIO model.

Authors and Affiliations

Omolaye O. Philip1 ;, Badmos T. Adeleke

Keywords

Related Articles

Feature Selection Optimization Solar Insolation Prediction Using Artificial Neural Network: Perspective Bangladesh

This paper presents a new climatological feature selection model using artificial neural network by utilizing the real world data for Bangladesh, called as FSOSIP. In this analysis back propagation algorithm is applied....

About Tenth Problem of D. Hilbert

In this article the author prove unsolvability of tenth problem of Hilbert

Role of Knit Factory Merchandiser for Efficient Supply Chain Management

Retailers face many challenges: time-to-market reductions are necessary due to shorter and shorter product life cycles, greater product variety causing more fluctuation in demand calls for high responsiveness in supply c...

Development of Mathematical Model for Cooling of Internal Combustion Engines

The energy released in the combustion chamber of an internal combustion engine is dissipated in three different ways. About 35 % of the fuel energy is converted to useful crankshaft work, and about 30 % energy is expelle...

Research Program of International Geophysical Year (Igy-1957) Have Ended In the Millennium Beginning By Discovery of the Theoretical Current Sheets in the Nature

We declare about the discovery in the Nature of theoretical current sheets by S. I. Syrovatskij. The first “natural” current sheet was discovered in flare spectrum, calculated and researched experimentally. The current s...

Download PDF file
  • EP ID EP401934
  • DOI -
  • Views 68
  • Downloads 0

How To Cite

Omolaye O. Philip1 ;, Badmos T. Adeleke (2017). Predictive and Comparative Analysis of NARX and NIO Time Series Prediction. American journal of Engineering Research, 6(9), 155-165. https://europub.co.uk./articles/-A-401934