Date of Award

Fall 2004

Document Type

Thesis

Degree Name

Master of Science in Computer Engineering - (M.S.)

Department

Electrical and Computer Engineering

First Advisor

Sirin Tekinay

Second Advisor

Cem U. Saraydar

Third Advisor

Roberto Rojas-Cessa

Abstract

The scalability problem of routing algorithms in Mobile Ad-hoc networks (MANET) has conventionally been addressed by introducing hierarchical architectures, clusters, and neighborhood zones. In all of these approaches, some nodes are assigned different routing related roles than others. Examples include cluster heads, virtual backbones and border nodes. The selection of these nodes on a fixed or dynamic basis adds complexity to the routing algorithm, in addition to placing significant demands on mobility and power consumption of these nodes. Furthermore, the scalability achieved with hierarchical architectures or partitions is limited.

This thesis demonstrates that location awareness can greatly aid in MANET routing and proposes an enhancement to location management algorithm used by the Terminodes System. This thesis makes use of geographic packet forwarding, geocasting and virtual home area concepts. It draws from the analogy between ad hoc networks and social networks. The Scalable Wide Area ad hoc network (SWAN), nodes update their location information with a geocast group whose area is given by a well-known function. A source node queries the geocast group of the destination and obtains up to date location information. Then, packets are geographically routed to the destination. The SWAN algorithm also optimizes the control overhead and obtains location information with minimal delay. This thesis also presents the results of our comparative performance study.

Share

COinS