The Ethernet Passive Optical Networks (EPONs) have been considered as a promising candidate for the next generation wired access networks for quite some time. In EPONs bandwidth requests and bandwidth allocations are critical issues which need to be addressed efficiently in order to guarantee the End-to-End (ETE) Quality of Service (QoS) for diverse classes of services. In this paper, we discuss the application of a statistical prediction technique based on the Bayesian Belief Networks (BBNs) which provides real-time decision support to the EPONs Dynamic Bandwidth Allocation (DBA) function. We show that in situations of burst traffic it helps the DBA to predict additional bandwidth requirements.
|Title of host publication||Unknown Host Publication|
|Number of pages||5|
|Publication status||Published - May 2011|
|Event||International Conference on Wireless and Optical Communications - Zhengzhou, China|
Duration: 1 May 2011 → …
|Conference||International Conference on Wireless and Optical Communications|
|Period||1/05/11 → …|
Moradpoor, N., Bashar, A., Parr, G., McClean, S., Scotney, B., & Owusu, G. (2011). Using Bayesian Belief Networks for Burst Detection in Ethernet Passive Optical Networks. In Unknown Host Publication IEEE.