Date of Award

Fall 2005

Document Type

Dissertation

Degree Name

Doctor of Philosophy in Electrical Engineering - (Ph.D.)

Department

Electrical and Computer Engineering

First Advisor

MengChu Zhou

Second Advisor

Nirwan Ansari

Third Advisor

Roy R. You

Fourth Advisor

Swades K. De

Fifth Advisor

I-Jy Steven Chien

Abstract

Advances in wireless technology and portable computing along with demands for high user mobility have provided a major promotion toward the development of ad hoc networks. Mobile ad hoc networks feature dynamic topology, self-organization, limited bandwidth and battery power of a node. They do not rely on specialized routers for path discovery and traffic routing. Research on ad hoc networks has been extensively investigated in the past few years and related work has focused on many of the layers of the communications architecture.

This research intends to investigate applications of MANET for intelligent systems, including intelligent transportation system (ITS), sensor network and mobile intelligent robot network, and propose some approaches to topology management, link layer multiple access and routing algorithms. Their performance is evaluated by theoretical analysis and off-the-shelf simulation tools.

Most current research on ad hoc networks assumes the availability of IEEE 802.11. However, the RTS/CTS protocol of 802.11 still leads to packet collision which in turn decreases the network throughput and lifetime. For sensor networks, sensors are mostly battery operated. Hence, resolving packet collision may improve network lifetime by saving valuable power. Using space and network diversity combination, this work proposes a new packet separation approach to packet collision caused by masked nodes.

Inter-vehicle communication is a key component of ITS and it is also called vehicular ad hoc network. VANET has many features different from regular MANETs in terms of mobility, network size and connectivity. Given rapid topology changes and network partitioning, this work studies how to organize the numerous vehicular nodes and establish message paths between any pair of vehicular nodes if they are not apart too far away.

In urban areas, the inter-vehicle communication has different requirements and constraints than highway environments. The proposed position-based routing strategy for VANETs utilizes the traffic pattern in city environments. Packets are forwarded based on traffic lights timing sequence and the moving direction of relaying vehicles. A multicast protocol is also introduced to visualize the real time road traffic with customized scale. Only vehicles related to a source node's planned trajectory will reply the query packet. The visualized real time traffic information therefore helps the driver make better decision in route planning when traffic congestion happens.

Nowadays robots become more and more powerful and intelligent. They can take part in operations in a cooperative manner which makes distributed control necessary. Ad hoc robot communication network is still fresh field for researchers working on networking technology. This work investigates some key issues in robot ad hoc network and evaluate the challenges while establishing robot ad hoc networks.

Share

COinS