DocumentCode :
1788827
Title :
A multi-agent EPON emulator
Author :
Sadon, S.K. ; Din, N.M. ; Radzi, N.A. ; Yaacob, M. ; Rawi, N.I.M. ; Maier, Martin ; Al-Mansoori, M.H.
Author_Institution :
Center for Commun. Service Convergence Technol., Univ. Tenaga Nasional, Kajang, Malaysia
fYear :
2014
fDate :
2-4 Sept. 2014
Firstpage :
62
Lastpage :
65
Abstract :
Passive Optical Networks (PONs) are widely used for the Fiber-to-the-Home (FTTH) solutions. Next generation PONs provides the platform for better service delivery at the access network. Managing the PON´s bandwidth is important to satisfy the users with different requirements and to ensure quality of service (QoS) while providing a well-managed network with minimum loss of bandwidth. It can be further optimized by making it more intelligent in providing service differentiation for different requirements based on a distributed approach. This paper presents an Ethernet-based PON emulator for optimizing bandwidth usage in the upstream environment where users need to share bandwidth based on an access mechanism. EPON emulator software is developed based on real time traffic data for EPON to validate different decentralized Dynamic Bandwidth Allocation (DBA) algorithms using Prolog. Prolog is an Artificial Intelligence based computer language, provides a platform for allowing ease of programming of the DBA using natural like language.
Keywords :
artificial intelligence; local area networks; passive optical networks; quality of service; Ethernet-based PON emulator; Fiber-to-the-Home solutions; Prolog; artificial intelligence based computer language; bandwidth usage; dynamic bandwidth allocation algorithms; multiagent EPON emulator; passive optical networks; quality of service; upstream environment; Bandwidth; EPON; IEEE 802.3 Standards; Load modeling; Optical network units; Passive optical networks; Real-time systems; DBA; EPON; Emulator; Multi-Agents;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Photonics (ICP), 2014 IEEE 5th International Conference on
Conference_Location :
Kuala Lumpur
Type :
conf
DOI :
10.1109/ICP.2014.7002311
Filename :
7002311
Link To Document :
بازگشت