Adaptive Scheduling Algorithms for Time-Slot Assignment in Distributed Radio Networks
Advisor: Prof. Harlan Russell
A cellular radio-network architecture is naturally suited to support multiple types of user traffic, including traffic such as voice and video that has tight constraints on the delay incurred in transmission through the network. However, there are many circumstances for which wireless, mobile communications is necessary but for which no supporting cellular infrastructure is available. For example, a group of laptop-computer users may wish to exchange data between their devices in a setting in which a fixed-site communications infrastructure is absent or not easily accessible. Or a rapid-deployment disaster-recovery team may require wireless communications capability in a region in which the communications infrastructure has been destroyed. A solution to this problem is for the mobile-communication devices to organize themselves into a network in which no device has centralized control (a peer-to-peer network). If some pairs of devices are not within radio-communication range of each other, the network must be capable of routing information through intermediate devices between a source and a destination (a multihop peer-to-peer network).
To satisfy the delay constraints, voice and video transmissions must be free from conflict with other interfering transmissions in the network with high probability. This is a very challenging requirement for a mobile, multihop, peer-to-peer network, but an approach that holds great promise for meeting this requirement is the use of a distributed scheduling algorithm in which radios are assigned transmission time slots designed for conflict-free transmissions in the network. The student investigator will evaluate the performance of a distributed scheduling algorithm (the Acyclic Slot Access Protocol algorithm) which has been developed at Clemson for use with a multihop, peer-to-peer radio network. The student will also develop and evaluate alternative approaches to adapting the transmission schedule in response to mobility in the network. The research will be focused on techniques that are appropriate for use with spread-spectrum radio networks. As part of the project, the student will learn the concepts and operation of distributed algorithms, learn the principles of stochastic modeling of networks, learn how to use simulation as a research tool, and learn the basic concepts of spread-spectrum communications.
Wireless Communications Program
home -- calendars -- campus map -- phone book -- search -- site index -- web accessibility
Clemson University, Clemson, South Carolina 29634 -- Area Code 864 -- Information 656-3311
Copyright © 2001, Clemson University. All rights reserved.