Core Network Function Placement in Self-Deployable Mobile Networks - Inria - Institut national de recherche en sciences et technologies du numérique
Article Dans Une Revue Computer Communications Année : 2019

Core Network Function Placement in Self-Deployable Mobile Networks

Résumé

Emerging mobile network architectures (e.g., aerial networks, disaster relief networks) are disrupting the classical careful planning and deployment of mobile networks by requiring specic self-deployment strategies. Such networks, referred to as self-deployable, are formed by interconnected rapidly deployable base stations that have no dedicated backhaul connection towards a traditional core network. Instead, an entity providing essential core network functionalities is co-located with one of the base stations. In this work, we tackle the problem of placing this core network entity within a self-deployable mobile network, i.e., we determine with which of the base stations it must be co-located. We propose a novel centrality metric, the ow centrality, which measures a node capacity of receiving the total amount of ows in the network. We show that in order to maximize the amount of exchanged trac between the base stations and the core network entity, under certain capacity and load distribution constraints, the latter should be co-located with the base station having the maximum ow centrality. We rst compare our proposed metric to other state of the art centralities. Then, we highlight the signicant trac loss occurring when the core network entity is not placed on the node with the maximum ow centrality, which could reach 55% in some cases.
Fichier principal
Vignette du fichier
comcom_oueis.pdf (2.28 Mo) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01928552 , version 1 (20-11-2018)

Identifiants

Citer

Jad Oueis, Vania Conan, Damien Lavaux, Hervé Rivano, Razvan Stanica, et al.. Core Network Function Placement in Self-Deployable Mobile Networks. Computer Communications, 2019, 133, pp.12-23. ⟨10.1016/j.comcom.2018.10.009⟩. ⟨hal-01928552⟩
203 Consultations
317 Téléchargements

Altmetric

Partager

More