A novel adaptive and distributed fair scheduling (ADFS) scheme for wireless sensor networks is shown through hardware implementation. In contrast to simulation, hardware evaluation provides valuable feedback to protocol and hardware development process. The proposed protocol focuses on quality-of-service (QoS) issues to address flow prioritization. Thus, when nodes access a shared channel, the proposed ADFS allocates the channel bandwidth proportionally to the weight, or priority, of the packet flows. Moreover, ADFS allows for dynamic allocation of network resources with little added overhead. Weights are initially assigned using user specified QoS criteria. These weights are subsequently updated as a function of delay, enqueued packets, flow arrival rate, and the previous packet weight. The back-off interval is also altered using the weight update equation. The weight update and the back-off interval selection ensure that global fairness is attained even with variable service rates. The algorithm is implemented using UMR/SLU motes for an industrial monitoring application. Results the hardware implementation demonstrates improved performance in terms of fairness index, flow rate, and delay.
M. J. Zawodniok et al., "Adaptive Distributed Fair Scheduling and Its Implementation in Wireless Sensor Networks," Proceedings of the IEEE International Conference on Systems, Man and Cybernetics, 2006, SMC'06, Institute of Electrical and Electronics Engineers (IEEE), Jan 2006.
The definitive version is available at https://doi.org/10.1109/ICSMC.2006.384641
IEEE International Conference on Systems, Man and Cybernetics, 2006, SMC'06
Electrical and Computer Engineering
Air Force Research Laboratory (Wright-Patterson Air Force Base, Ohio)
United States. Department of Education
University of Missouri--Rolla. Intelligent Systems Center
Keywords and Phrases
Adaptive-Fair-Scheduling; Quality-Of-Service; Weight-Adaptation; Embedded computer systems; Fairness
Article - Conference proceedings
© 2006 Institute of Electrical and Electronics Engineers (IEEE), All rights reserved.