The communications spanning tree problem: Comparison of an heuristic random search methods and lower bounding
Given a set of nodes and a list of offered traffic between OD pairs a communication tree is a spanning tree such that each link has sufficient capacity to carry all traffic assigned to it. In this paper we discuss several approaches for finding minimum cost communication spanning trees, where the cost of the tree is the sum of the link costs. Each link cost is taken to be the product of the length of the link and its capacity. This problem is NP-complete and so there is probably no efficient algorithm with which to solve it. The paper will describe three solution methods - an heuristic, simulated annealing and a genetic algorithm. We will then look at a lower bounding procedure using lagrangian relaxation and present the results of extensive computational experience.
- OSTI ID:
- 35864
- Report Number(s):
- CONF-9408161--
- Country of Publication:
- United States
- Language:
- English
Similar Records
Improving spanning trees by upgrading nodes
2-D image segmentation using minimum spanning trees