Convex optimization over fixed point sets of quasi-nonexpansive and nonexpansive mappings in utility-based bandwidth allocation problems with operational constraints
-
Hideaki Iiduka [1]
-
[1] Meiji University
Meiji University
Japón
-
- Localización: Journal of computational and applied mathematics, ISSN 0377-0427, Vol. 282, Nº 1 (July 2015), 2015, págs. 225-236
- Idioma: inglés
- DOI: 10.1016/j.cam.2014.12.048
- Enlaces
-
Resumen
Network bandwidth allocation is a central issue in modern communication networks. The main objective of the bandwidth allocation is to allocate an optimal bandwidth for maximizing a predefined utility over the capacity constraints to traffic sources. When a centralized operator, which manages all the bandwidth allocations in the network, has a certain operational policy, the bandwidth allocation reflecting the operational policy should result in the network being more stable and reliable. Accordingly, we need to solve a network bandwidth allocation problem under both capacity constraints and operational constraints. To develop a novel algorithm for solving the problem, we translate the network bandwidth allocation problem into one of minimizing a convex objective function over the intersection of the fixed point sets of certain quasi-nonexpansive and nonexpansive mappings and propose a fixed point optimization algorithm for solving it. We numerically compare the proposed algorithm with the existing algorithm for solving a concrete bandwidth allocation problem and show its effectiveness.
