Abstract

In general, humans follow highly predictable daily movements. We commute from residential to working and/or educational areas in a daily basis, and we have a selection of commercial and recreational areas for the nights and weekends. We also use the mobile phone at regular hours, for example when commuting, during lunch break and at night. Such regular behavior creates predictable spatio-temporal fluctuations of traffic patterns, which in analogy to the periodical rise and fall of the sea levels, its known as tidal traffic phenomenon. In this paper, we propose a suite of on-line matheuristics to reduce energy consumption of optical layer in mobile metro-core networks, while providing 1 + 1 protection of the aggregated traffic. The proposed matheuristic dynamically optimizes resource allocation by effectively adapting to predictable tidal traffic variations. The optimality of on-line decisions is achieved by generating optimal weighted graphs from the interaction with an off-line optimization phase. In the off-line phase either a wavelength path (WP) or a virtual-WP problem is solved using the regular tidal traffic patterns. Our results display energy savings of more than 20% by introducing load adaptive network operation, while the matheuristic closely follows the optimal results. We also introduce a heuristic method to reduce service disruption due to routing changes, while preserving energy saving capability. The heuristic allows us to reduce service disruption in the range of 38–80% with a small penalty on energy saving of less than 5%.

© 2016 IEEE

PDF Article

References

You do not have subscription access to this journal. Citation lists with outbound citation links are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access OSA Member Subscription

Cited By

You do not have subscription access to this journal. Cited by links are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access OSA Member Subscription