Title: Prediction of wireless network connectivity using a Taylor Kriging approach

Authors: Heping Liu, Soroor K. Al-Khafaji, Alice E. Smith

Addresses: Department of Industrial and Systems Engineering, Auburn University, Auburn, AL, 36849, USA. ' Department of Industrial and Systems Engineering, Auburn University, Auburn, AL, 36849, USA. ' Department of Industrial and Systems Engineering, Auburn University, Auburn, AL, 36849, USA

Abstract: The research aim of this paper is to investigate the effectiveness of a new Kriging model which uses Taylor expansion to predict wireless network connectivity. Wireless network connectivity is measured by the strength of emitted signal power from the tower to the point in question. The prediction results are compared with those from the literature where an Ordinary Kriging model and a neural network are used to conduct the same prediction. Root mean squared error (RMSE) and maximum absolute relative error (MARE) show that the prediction results of the new Kriging model are much better than those obtained before with average differences from 51.56% to 85%. This study shows the promise of the new Kriging model to accurately estimate wireless signal strength.

Keywords: Taylor Kriging; prediction modelling; wireless networks; network connectivity; signal strength; wireless signals.

DOI: 10.1504/IJAIP.2011.039744

International Journal of Advanced Intelligence Paradigms, 2011 Vol.3 No.2, pp.112 - 121

Published online: 30 Sep 2014 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article