Optimal Placement of Add/Drop Multiplexers: Heuristic and Exact Algorithms

Authors:
Alain Sutter;Francois Vanderbeck;Laurence Wolsey
Affiliations:
-;-;-
Venue:
Operations Research
Year:
1998

Citing 0
Cited 8

Splittable traffic partition in WDM/SONET rings to minimize SONET ADMs

Theoretical Computer Science
Bi-directional self-healing ring network planning

Computers and Operations Research
An Evolutionary Algorithm for the Design of Hybrid Fiber Optic-Coaxial Cable Networks in Small Urban Areas

IWANN '01 Proceedings of the 6th International Work-Conference on Artificial and Natural Neural Networks: Connectionist Models of Neurons, Learning Processes and Artificial Intelligence-Part I
SONET/SDH ring assignment with capacity constraints

Discrete Applied Mathematics - Special issue: Algorithmic aspects of communication
On stacking bi-directional self-healing-rings on a conduit ring

Computers and Industrial Engineering
A ring-mesh topology design problem for optical transport networks

Journal of Heuristics
Cutting plane algorithms for solving a stochastic edge-partition problem

Discrete Optimization
Optimal placement of reconfigurable optical add/drop multiplexers with packing, blocking, and signal loss

Computer Networks: The International Journal of Computer and Telecommunications Networking

Quantified Score

Hi-index	0.00

Visualization

Abstract

We study a problem that has arisen recently in the design of telecommunications transmission networks at France Telecom. Given a set of centers in a city or conglomeration linked together on a ring architecture, given the expected demands between the centers and an essentially unlimited availability of rings of fixed capacity on the network, assign demand pairs and corresponding add/drop multiplexers to the rings so as to satisfy the demands and minimize the number of "costly" multiplexers installed. Heuristics based on simulated annealing have been developed for the basic problem and several variants. France Telecom is particularly interested in validating the effectiveness of the heuristics. An exact algorithm based on integer column generation is shown to provide tight performance guarantees, and in most cases to give provable minimum cost solutions. Column generation is used even though the subproblems are strongly NP-hard and are solved by standard mixed integer programming software. In addition, the Master Problems involve both integer columns and integer variables, whereas these are 0-1 in most reported applications.