Bandwidth optimization by reliable-path determination in Mobile Ad-Hoc Networks

  • Authors:
  • Sriram Raghavan;Srikanth Akkiraju;Santhosh Sridhar

  • Affiliations:
  • Anna University;Anna University;Anna University

  • Venue:
  • Proceedings of the 3rd international conference on Mobile and ubiquitous multimedia
  • Year:
  • 2004

Quantified Score

Hi-index 0.00

Visualization

Abstract

A Mobile Ad Hoc Network is typically characterized by two sets of nodes, one set being more stationary than the other. It is observed that identifying this set of relatively stationary nodes critically facilitates the dual purpose of reducing the routing overhead and providing larger data bandwidth. It is therefore both interesting and expedient to study the behavior of a typical Mobile Ad Hoc Network and identify the sub-graph which can impact these improvements where the application demands are high. It is particularly enlightening when scaled to a larger network, as one of the substantial factors affecting cost in an Ad Hoc network is the routing overhead. In this paper, a method is proposed to determine, dynamically, the most reliable path for routing in a Mobile Ad Hoc scenario and ensure that packets are routed along that path to guarantee enhanced data traffic routed per unit time owing to savings from route-discovery and to decrease normalized routing overhead. The Mobile Ad Hoc Network is modeled as a graph with the mobile units denoting the nodes of the graph and the set of reachable units as its neighbors and identifying the weakly connected sub-dominion of that graph, which is the most reliable path for routing. The proposed algorithm recognizes those set of weakly connected nodes which are relatively more stable when compared to the other nodes hence requiring less routing and thus resulting in more number of data packets and ensuring better bandwidth utilization. Our algorithm proposes and confirms the dual betterments namely improvement in the utilization of bandwidth and reduction in the normalized routing overhead in any given network scenario.