A simple neural framework for bandwidth reservation of VoIP communications in cost-effective devices

Author

Caviglione, Luca

Author_Institution

Inst. for Intell. Syst. for Autom. (ISSIA) Genoa Branch, Italian Nat. Res. Council (CNR), Genoa, Italy

Volume

56

Issue

3

fYear

2010

Firstpage

1252

Lastpage

1257

Abstract

Voice over IP (VoIP) is a widely recognized killer application of the Internet and it has been integrated within many services (for instance to enrich the on-line gaming experience). In addition, such feature is often natively supported by mobile devices equipped with IEEE 802.11-based communications, or by standard UMTS (Universal Mobile Telecommunication System) cellular phones. Nevertheless, Beyond 3rd Generation (B3G) telephony is expected to fully support IP-based communications. In this perspective, this paper analyzes the feasibility of adopting a simple framework, based on neural networks, to reserve bandwidth, when in presence of VoIP communications. The framework has been designed to be simply embeddable within a wide range of cost-effective consumer devices (e.g., xDSL modems, cellular phones and network appliances).

Keywords

3G mobile communication; Internet telephony; neural nets; Beyond 3rd Generation telephony; IEEE 802.11; Internet; UMTS; Universal Mobile Telecommunication System; VoIP communications; bandwidth reservation; cellular phones; cost-effective consumer devices; neural networks; voice over IP; xDSL modems; Artificial neural networks; Bandwidth; Computer architecture; Protocols; Quality of service; Throughput; Training; VoIP, Bandwidth Reservation, Neural Network, Black Box Modeling, QoS;

fLanguage

English

Journal_Title

Consumer Electronics, IEEE Transactions on

Publisher

ieee

ISSN

0098-3063

Type

jour

DOI

10.1109/TCE.2010.5606255

Filename

5606255