Design guidelines for quality of service support in Optimized Link State Routing-based mobile ad hoc networks

  • Authors:

  • P. Sondi;D. Gantsou;S. Lecomte

  • Affiliations:

  • Univ. Lille Nord de France, F-59000 Lille, France and UVHC, LAMIH-DIM, F-59313 Valenciennes, France and CNRS, UMR 8201, F-59313 Valenciennes, France;Univ. Lille Nord de France, F-59000 Lille, France and UVHC, LAMIH-DIM, F-59313 Valenciennes, France and CNRS, UMR 8201, F-59313 Valenciennes, France;Univ. Lille Nord de France, F-59000 Lille, France and UVHC, LAMIH-DIM, F-59313 Valenciennes, France and CNRS, UMR 8201, F-59313 Valenciennes, France

  • Venue:
  • Ad Hoc Networks
  • Year:
  • 2013

Quantified Score

Hi-index 0.00

Visualization

Abstract

The proposals for quality of service in mobile ad hoc networks have focused on adding QoS mechanisms in best-effort routing protocols while keeping the same assumptions regarding both the network representation and the routing protocol design. Most of the solutions concentrate the algorithmic effort on QoS route computation and relax the constraint of optimizing flooding, which has, however, a great impact on resource consumption. The goal of this paper is to present guidelines that allow the design of efficient solutions accurately presented and easily comparable with each other. We first define the QoS metric as an abstraction that can be instantiated for any kind of metric, and we propose a MANET representation that integrates this abstraction into the definition of the network graph and properties. Using these concepts, while flooding optimization and QoS-aware MPR selection have usually been performed separately, we then propose a formalism that unifies MPR selection in such a way that both objectives are achieved simultaneously in a single selection. Finally, we propose a heuristic that provides an efficient solution for this problem while allowing us to control the trade-off between both objectives. Evaluations carried out on network graphs considering different spatial distributions of nodes show that our heuristic outperforms existing heuristics.