A Novel Fault Diagnosis Approach to Path-Protected WDM Mesh Networks

  • Authors:
  • Chi-Shih Chao

  • Affiliations:
  • Department of Communications Engineering, Feng Chia University, Taiwan, ROC 40724

  • Venue:
  • APNOMS '08 Proceedings of the 11th Asia-Pacific Symposium on Network Operations and Management: Challenges for Next Generation Network Operations and Service Management
  • Year:
  • 2008

Quantified Score

Hi-index 0.00

Visualization

Abstract

In today's cyber-world, network faults can not only cause unexpected damages and consternation to enterprises, but also result in huge loss of customers and a lot of negative effects. For this reason, ISPs (Internet Service Providers) do their best to achieve the goal of higher network reliability, service availability, and accessible bandwidth. To do so, most of them construct or rent carrier-grade backbone networks, which use WDM (Wavelength Division Multiplexing) with path restoration/protection techniques, as one of the means toward the goal. However, network faults can still impact largely to these wellprotected WDM networks since the actual network reliability would fall down as a fault occurs, even if the network could function as normal with protection mechanisms. In addition, the protection/restoration mechanisms would usually get fault diagnosis into trouble because they can cover faults in most cases. In this paper, a novel and competent fault diagnosis approach which can be used in path-protected WDM mesh networks for a high and constant network availability is proposed. This diagnosis approach uses the trimmed/extracted major alarm propagation behavior of a fault and our leader major alarm domain concept to deduce and isolate the most suspicious network area of the fault. Additionally, to enhance the diagnosis accuracy, our approach also refers to some certain information with respect to SRLG (Shared Risk Link Group) and dynamic network sessions of the diagnosed network to further narrow down the suspicious area. Last, simulation results associated with discussions are shown to demonstrate the performance of our diagnosis approach.