CAREER: Energy-Efficient Protocols for Multimedia Traffic in Ad Hoc Networks

Clemson University, Clemson SC

Investigators

Abstract

Support for multiple quality-of-service (QOS) levels in mobile, distributed, packet radio networks is required for many commercial wireless applications. Mobile radio networks (or ad hoc networks) are required in situations where the network itself must be highly mobile and where access to a fixed infrastructure of base stations is not available. To meet demands of portability and low-power operation, it is desirable for a single ad hoc network to service the traffic demands of a wide range of applications, including examples such as all-informed multicast voice, real-time video, large file transfers, and interactive web browsing. Each class of traffic has specific service requirements and these QOS levels must be provided from a mobile and distributed network that operates with battery power only and with very limited bandwidth. The Wireless Communications Program at Clemson University has recognized that strong interaction among different protocol layers in ad hoc radio networks is not only desirable but in fact essential. A single protocol layer cannot guarantee a specific quality of service for a traffic type, but there must be coordination among the routing, queuing, traffic scheduling, congestion control, channel-access, and adaptive-transmission protocols. We have already demonstrated that information from the physical and link layers can be effectively utilized by the routing protocol to dramatically improve the end-to-end network performance and reduce the total energy expended by the mobile radios. The proposed research and education plan is based on our previous results that show that strong coupling among the protocol layers is vital for wireless communications networks that must interconnect highly mobile users and link them reliably to a fixed infrastructure. This career plan is structured to extend previous results through collaborative research that is focused on the link and network layers, and to explore a new topic area by applying the advances discovered in our previous work to new research into the interaction between the network and transport layers. The proposed career plan is centered on both undergraduate and graduate student participation. The topics are organized so that students with a range of experience levels can make significant contributions. A key priority of the research plan is to include students with a variety of career goals and motivate them to strive for the highest levels of scientific and engineering excellence throughout their careers. Ample opportunities are planned to encourage and develop promising undergraduate students to pursue graduate school. An important component of the educational plan is to incorporate research tools in the classroom. Animation tools that illustrate the performance of our distributed protocols have increased our understanding our research and they will prove to be equally valuable in the classroom and laboratory. Rather than simply reading about networking protocols in a textbook, students will implement distributed protocols within a discrete event simulation program and see the performance implications of the various protocols.