Title: Neural modelling and performance analysis of the link occupancy distribution for wireless broadband transmissions

Authors: Izabella Lokshina, Michael R. Bartolacci

Addresses: Management, Marketing and Information Systems, SUNY Oneonta, Oneonta, NY 13820, USA. ' Information Sciences and Technology, Penn State University, Berks, PA 19610, USA

Abstract: This paper is devoted to neural modelling and performance analysis of the link occupancy distribution for wireless broadband transmission. Multiclass models of a single link transmission for rigid, adaptive and elastic traffic are developed, based on Markov rewards models. The link occupancy distribution is introduced as embedded, discrete time Markov chains researched with the use of Vector Quantification (VQ). Link occupancy performance is simulated as a combination of single queues with random distributions of arrival processes and holding time service phases. The density of occupancy probability is determined using Learning Vector Quantification (LVQ) in a two-layered neural structure.

Keywords: artificial intelligence; neural networks; Markov reward models; discrete time Markov chains; communication networks; performance analysis; link occupancy distribution; wireless networks; broadband transmissions; mobile networks; neural modelling.

DOI: 10.1504/IJMNDI.2007.015063

International Journal of Mobile Network Design and Innovation, 2007 Vol.2 No.2, pp.129 - 133

Published online: 06 Sep 2007 *

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