Distance-adaptive, Low CAPEX Cost p -Cycle Design Without Candidate Cycle Enumeration in Mixed-Line-Rate Optical Networks

Even though elastic optical networks (EONs) are promising to provision increasingly dynamic and heterogeneous traffic, the requirements on bandwidth-variable optical devices bring upgrading challenges in current wavelength-division multiplexing (WDM) optical networks. Mixed-line-rate (MLR) optical n...

Full description

Saved in:
Bibliographic Details
Published in:Journal of lightwave technology 2016-06, Vol.34 (11), p.2663-2676
Main Authors: Ju, Min, Zhou, Fen, Zhu, Zuqing, Xiao, Shilin
Format: Article
Language:English
Subjects:
Algorithm design and analysis
Algorithms
Computational modeling
Computer simulation
Cost engineering
Devices
Distance-adaptive
Enumeration
Expenditures
Mixed Integer Linear Programming (MILP)
Mixed Line Rates (MLR) Optical Networks
Optical communication
Optical network units
Partitioning algorithms
Pre-Configured-Cycle (p-Cycle)
Traffic flow
Transponders
Wavelength division multiplexing
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Even though elastic optical networks (EONs) are promising to provision increasingly dynamic and heterogeneous traffic, the requirements on bandwidth-variable optical devices bring upgrading challenges in current wavelength-division multiplexing (WDM) optical networks. Mixed-line-rate (MLR) optical networks offer a transitional solution that allows several coexisting line rates (e.g., 10/40/100 Gb/s). In this paper, we investigate distance-adaptive preconfigured-cycle (pCycle) protection scheme in MLR optical networks. Path-length-limited p-cycles are designed to be assigned line rate depending on the length of each protection path. Instead of conventional candidate cycle enumeration, a mixed integer linear programming (MILP) model is formulated to directly generate the optimal p-cycles with the minimum capital expenditures (CAPEX) cost. We also develop two algorithms to make the proposed MILP model scalable. Simulation results indicate that the algorithms are time efficient for solving the MILP-based p-cycle design. We further compare our p-cycle design method with other schemes, and demonstrate that our method largely reduces the CAPEX cost for more than 40%, mainly in transponder cost. To the best of our knowledge, this is the first time that distance-adaptive p-cycle design without candidate cycle enumeration is proposed for MLR optical networks.
ISSN:0733-8724
1558-2213
DOI:10.1109/JLT.2016.2549642