Performance evaluation of p-cycle based optical multicast protection techniques.
Wong, Kelvin Teck Koon.
Date of Issue2009
School of Electrical and Electronic Engineering
In today’s high capacity wavelength division multiplexed (WDM) optical networks, the failure of a network component, such as a fiber link, could lead to the failure of all lightpaths traversing that particular link. Hence, it is especially important to design WDM optical networks that are inherently fault‐tolerant against events like fiber cuts, central office and hardware failures. A survivable WDM optical network is able to provide alternate routes between a pair of nodes when a failure occurs. The primary objective of network survivability is to maintain the end to end data service in the event of any network element failures such as link or node failures. Survivability in optical network involves detecting any network element failure and performing a protection switching to divert the traffic through an alternate path when the failure occurs. The two basic approaches for network survivability are protection and restoration. The protection method pre‐determines backup paths and reserve resources at the same time when the corresponding working lightpaths are established. The restoration method on the other hand searches for a backup path only when a failure occurs in the network. The protection method can guarantee the full recovery at fast speed while restoration method may outperform the former in resource utilization efficiency. However, the drawback of the restoration method is that it cannot guarantee full recovery and may take a much longer time for recovery due to the signalling time. One of most significant recent developments in survivable network design is the p‐cycle protection method, with ring‐like speed and mesh‐like efficiency. In this report, our emphasis is on the protection approach for survivability in the optical layer, in particular, the pre‐configured protection cycle (p‐cycle) method. The p‐cycle method can provide ring‐like fast protection and mesh‐like high efficiency in the use of spare capacity. An extension of the p‐cycle protection method is the path‐segment‐protecting p‐cycle (flow p‐cycle). This particular approach takes an explicit view of the service path crossing the network and how they relate to the p‐cycle we use for protection. Flow p‐cycle as previous results have demonstrated, its capacity efficiency is near that of the shared backup path‐protection (SBPP) scheme currently favoured for optical networking. Current approaches of p‐cycle based optical multicast protection technique only takes into account protection of a single source to multiple destinations. In this case, when the source encounters a node failure, the p‐cycle protection cannot protect the network at all, resulting in end to end data service failure. In view of this, the author proposes a p‐cycle protection method against single source node failure with the introduction of a secondary source. By applying the flow p‐cycle protection approach on a two source network, we are able to protect against primary node failure. The author compares this protection approach with the optimal pair‐path protection method and the results show that there is an average of 25% improvement in the capacity efficiency for all multicast groups. In this project, the author has fulfilled all the objectives required and also proposed a more efficient method to protect a given topology.
DRNTU::Engineering::Electrical and electronic engineering::Optics, optoelectronics, photonics
Final Year Project (FYP)
Nanyang Technological University