Abstract:
Optical networking is a promising solution to the growing bandwidth
requirements of today’s communication networks. The continuous challenge
is to look for technological advances to improve network capacity,
reliability, scalability and cost. This work addresses some of these factors by
presenting a rerouting strategy for reducing the effect of the wavelength
continuity constraint and improving the performance of all-optical
wavelength division multiplexing (WDM) networks in terms of blocking
probability. The problem of routing and wavelength assignment (RWA), of
main interest in this study, is critically important for increasing the efficiency
of WDM networks under wavelength continuity constraint.
The proposed scheme introduces a rerouting strategy based on wavelength
reassignment, with the First-Fit (FF) wavelength assignment algorithm, and
route deviation, utilizing the Least Congested Path (LCP) routing algorithm.
In addition, this approach is applied in the context of quality of service (QoS)
differentiation to help network service providers to satisfy and guarantee
service transmission priorities and tolerable delay requirements.
Finally, a neural network model is developed to investigate the prediction of
upcoming blocking times in the network in an attempt to reduce the network disruption caused by unnecessary rerouting, therefore improving the
performance of the proposed strategy. Extensive simulations allowed us to
verify our proposed rerouting strategy and enhancement models and assert
their effectiveness in enhancing the network performance in terms of
blocking probability especially for networks utilizing a small number of
wavelengths per fiber.
