Abstract

Available congestion control schemes, for example transport control protocol (TCP), when applied to wireless networks results in a large number of packet drops, unfairness with a significant amount of wasted energy due to retransmissions. To fully utilize the hop by hop feedback information, a suite of novel, decentralized, predictive congestion control schemes are proposed for wireless sensor networks in concert with distributed power control (DPC). Besides providing energy efficient solution, embedded channel estimator in DPC predicts the channel quality. By using the channel quality and node queue utilizations, the onset of network congestion is predicted and congestion control is initiated. Stability of the hop by hop congestion control is demonstrated by using a Lyapunov-based approach. Simulation results show that the proposed schemes result in fewer dropped packets than a network without the hop-by-hop congestion control, better fairness index and network efficiency, higher aggregate throughput, and smaller end-to-end delays over the other available schemes like IEEE 802.11 protocol.

Meeting Name

International Conference on Control and Automation, 2005. ICCA '05

Department(s)

Electrical and Computer Engineering

Second Department

Computer Science

Sponsor(s)

National Science Foundation (U.S.)

Keywords and Phrases

Lyapunov Methods; Access Protocols; Channel Estimation; Wireless Sensor Networks

Library of Congress Subject Headings

Predictive control
Queuing theory

Document Type

Article - Conference proceedings

Document Version

Final Version

File Type

text

Language(s)

English

Rights

© 2005 Institute of Electrical and Electronics Engineers (IEEE), All rights reserved.