Harlan B. Russell
Harlan Russell Associate Professor of Electrical and Computer Engineering
Ph.D., 1993 - University of Illinois, Urbana Champaign
M.S., 1989 - University of Illinois, Urbana Champaign
B.S., 1986 - University of Illinois, Urbana Champaign
Office: 316 Fluor Daniel EIB
Office Phone: 864.656.7214
Personal home page
From 1993 to 1999, Dr. Russell was a Senior Research Engineer for Techno-Sciences, Inc., Pendleton, SC. While at Techno-Sciences he worked on a number of joint projects with faculty at Clemson University and industrial partners ITT Industries and BBN Technologies. A GloMo project funded by DARPA is one such example, and consisted of adding advanced link and network layer protocols to ITT's Handheld Multimedia Terminal (HMT). The network protocol suite, called RAVEN, combines an adaptive routing protocol with a virtual circuit protocol to support data, voice, and video traffic in ITT's radio. He joined the faculty at Clemson University in December 1999 as an Assistant Professor.
Dr. Russell's research interests are in distributed protocols for wireless networks. Highly mobile communication networks are needed for applications such a military operations, emergency search-and-rescue, and data collection in inhospitable environments. These self-configuring networks are formed by the ad hoc collection of radios that are available at the time and location communication is required, and these networks should not depend on an existing infrastructure. His group is conducting research on link, network, and transport layer protocols that provide different quality-of-service levels for multimedia traffic in ad hoc networks.
Reliable and Efficient Mobile Wireless Networks
A challenge in the design of protocols for ad hoc networks is the unreliability of the wireless links. A novel aspect of our work employs information from the communication receiver, including the demodulator and decoder, to characterize the link quality. The link quality information is incorporated into the network protocols to improve the forwarding, routing, and transmission protocols. For example, we have introduced least-resistance routing, a routing protocol that can quickly adapt to dynamic network topologies. Routes are selected based on a number of different criteria such as the level of interference, the ability of the links to support different quality of service requirements for the traffic, and the energy required in successfully delivering traffic.
Shared Access to the Radio-Frequency Spectrum
The radios in an ad hoc network must share the radio-frequency spectrum fairly. We are investigating both contention-based and scheduled approaches to channel access. Contention-based channel-access protocols permit radios to begin transmissions as packets are available, but the protocols must negotiate when transmissions are permitted and how to recover from collisions due to conflicting demands. Our group is designing protocols that can dynamically adapt to the environment based on the level of demands and the density of transmitters to maximize spectral efficiency. Schedule channel-access protocols avoid collisions and provide for the possibility of very efficient access to the channel when there is a high demand for shared access to the spectrum. We are investigating protocols that jointly schedule transmissions and select routes for the relay of packets that maximize the network capacity and adapt to bottlenecks.
Protocols for Frequency-agile Radios
While some portions of the spectrum are becoming saturated due to the increasing demands for mobile and wireless applications, other bands are idle or underutilized. Emerging software-defined radios are enabling increasing flexibility to utilize multiple frequency channels with widely different characteristics. As components of the radios that have typically been implemented in hardware are replaced with software-based systems, it becomes possible to update the software quickly to change the transmission frequency and format with very little delay. These types of frequency-agile radios allow the channel parameters to be modified at the time the network is deployed or as the operating conditions change. Our group is investigation new protocols that permit the radios to self-configure to utilize these bands as they become available.