Development of Geocasting Protocols for a MANET

Colorado School Of Mines, Golden CO

Investigators

Abstract

Ad hoc networking involves computers, typically wireless mobile nodes, that cooperatively form a network without specific user administration or configuration. In other words, ad hoc networking allows an arbitrary collection of mobile nodes to create a network on demand. This capability has numerous applications in tactical missions and rescue operations, as well as for educational and commercial use. Due to the number of applications desiring the formation of an ad hoc network infrastructure, development of this technology is currently a research priority at NSF. The work in this proposal concerns the development and performance evaluation of protocols that offer geocast communication in an ad hoc network. The goal of a geocasting protocol is to deliver a packet to a set of nodes within a specified geographical area, i.e., the geocast region. The researchers will develop and evaluate protocols that offer geocast communication to both explicityly defined groups (i.e., geocast to those mobile nodes in the geocast region that have registered with the group) and implicitly defined groups (i.e., geocast to all mobile nodes in the geocast region). The researchers will also consider both fixed geocast regions (i.e., defined by a specific geographical location) and dynamic geocast regions (i.e., defined by a dangerous moving target). The researchers will develop and evaluate (via simulation) three different approaches to geocasting: a directed flooding approach, a tree-based approach, and a mesh-based approach. They also propose to develop a hybrid approach that is based on the nature of the message to be transmitted and/or performance considerations (e.g., if the mobile nodes have been moving slowly in the recent past, then the redundancy offered by a mesh-based approach may not be needed). Lastly, the researchers plan to incorporate predictable and controllable end-to-end quality of service when a shared tree or shared mesh approach is used. Throughout the project, they will conduct extensive simulations to determine the conditions under which each of the geocast algorithms is most effective. The researchers will vary the essential parameters in the simulations to avoid biased results, e.g., network size, connectivity, topological rate of change, and various movement/calling patterns. In addition, the simulations will take into account some realistic limitations that have not always been addressed in prior performance evaluations, e.g., location errors that exist in widely available consumer Global Positioning Systems. The proposed project is a collaboration between two investigators who have the experience and capability necessary to complete the project successfully. The expertise of one investigator is in the mobile computing and networking area. The expertise of the other investigator is in the development of simulation models and the evaluation fo a protocol's performance. Throughout the investigation, the approach will be from a practical sense. The researchers will attend Internet Engineering Task Force meetings to obtain the current status of ad hoc protocols and to share the results.