National Changhua University of Education Institutional Repository : Item 987654321/10560

English | 正體中文 | 简体中文 | Items with full text/Total items : 6507/11669
Visitors : 29934615 Online Users : 456

RC Version 3.2 © Powered By DSPACE, MIT. Enhanced by NTU Library IR team.

Scope Adv. Search

Login ‧ Upload ‧ Help ‧ About ‧ Administer

NCUEIR > College of Engineering > Department and Graduate Institute of Computer Science and Information Engineering > Periodical Articles > Item 987654321/10560

Please use this identifier to cite or link to this item: http://ir.ncue.edu.tw/ir/handle/987654321/10560

Title:	An Area-Based Vertical Motion Estimation on Heterogeneous Wireless Networks
Authors:	Chang, Ing-Chau;Wang, Ching-Hsiang;Chang, Lin-Huang
Contributors:	資訊工程系
Date:	2006-08
Issue Date:	2012-05-25T07:42:50Z
Publisher:	Springer Verlag
Abstract:	In this paper, we will propose an area-based vertical motion estimation (AVME) scheme to efficiently and accurately predict target wireless networks and corresponding target cells for the next handoff of the mobile node (MN) on the increasingly prevalent heterogeneous wireless network environment which consisting of several different wireless networks such as wireless LAN (WLAN), third generation cellular network (3G), etc. We adopt the back propagation neural network (BPN) model in this AVME scheme to generate mobility patterns of MNs by training the BPN model with historical handoff information of these MNs. By using the IBM City Simulator [1] to create the city plan and motions of all MNs, simulation results show that our AVME scheme on heterogeneous wireless networks can achieve higher level of predication accuracy for next handoff than traditional cell-based prediction scheme does on a single wireless network, but with less resources and computations.
Relation:	LNCS, 4138: 187-198
Appears in Collections:	[Department and Graduate Institute of Computer Science and Information Engineering] Periodical Articles

Files in This Item:

File	Size	Format
2050400510005.pdf	7Kb	Adobe PDF	602	View/Open