Summary
International Symposium on Nonlinear Theory and its Applications
2009
Session Number:A1L-D
Session:
Number:A1L-D5
Application QoS based Optimization of Heterogeneous Wireless Networks by Distributed and Autonomous Neural Network Dynamics
Akira Funatsu, Taichi Takeda, Taro Kuroda, Mikio Hasegawa , Homare Murakami, Hiroshi Harada,
pp.-
Publication Date:2009/10/18
Online ISSN:2188-5079
Summary:
Various radio access network (RAN) services have been developed and commercialized. In order to always use the best connection, vertical handover technologies for adaptively switching among those various different RANs without interruption of an ongoing communication session have been developed, and several protocols to enable such handover have been standardized. In this paper, we propose an optimal RAN selection algorithm, which uses autonomous and distributed neural network dynamics, to optimize radio resources usage and quality of services (QoS) in heterogeneous wireless network environment. We introduce a higher order neural network to optimize satisfaction rate of QoS level required by applications on the user terminals. By computer simulation, we show that the proposal method improves the QoS satisfaction rate by distributed update of neurons, which doesn’t require any centralized computation.