Autonomous routing algorithms for networks with wide-spread failures
Name
Khan-2009-Autonomous routing algorithms for networks with wide-spread failures.pdf
Size
4.43 MB
Format
Adobe PDF
Checksum (MD5)
51d11c4e93a5ee1bb02bcfaa2af85ff4
Author(s) • •
Modiano, Eytan H.
Khan, Wajahat F.
Le, Long Bao
Date Issued
January 2010
Journal
IEEE Military Communications Conference, 2009. MILCOM 2009
Publisher
Institute of Electrical and Electronics Engineers
Citation
Wajahat Khan, Long Bao Le, and E. Modiano. “Autonomous routing algorithms for networks with wide-spread failures.” Military Communications Conference, 2009. MILCOM 2009. IEEE. 2009. 1-6. ©2010 Institute of Electrical and Electronics Engineers.
Version
Final published version
Abstract
We study end-to-end delay performance of different routing algorithms in networks with random failures. Specifically, we compare delay performances of differential backlog (DB) and shortest path (SP) routing algorithms and show that DB routing outperforms SP routing in terms of throughput when the network is heavily loaded and/or the failure rate is high while SP routing achieves better delay performance in the low load regime. Then, we investigate delay performance of a hybrid routing algorithm that combines principles of both SP and DB routing algorithms and show that it outperforms both of these routing algorithms. Finally, we demonstrate improvements in delay performance of DB routing through the use of a digital fountain approach which was originally proposed for multicast applications. In addition, our results show that there exists an optimal coding rate where digital fountain based DB routing achieves minimum end-to-end delay. To the best of our knowledge, this is the first work which investigates delay performance of DB routing and its enhanced versions for networks with link failures.
Subjects
Differential backlog routing
digital fountain
end-to-end delay
shortest path routing
throughput region
MIT Department
Massachusetts Institute of Technology. Communications and Networking Research Group
Massachusetts Institute of Technology. Department of Aeronautics and Astronautics
Terms of Use
Article is made available in accordance with the publisher's policy and may be subject to US copyright law. Please refer to the publisher's site for terms of use.
Persistent DSpace Link
DOI of Published Version
https://doi.org/10.1109/MILCOM.2009.5379792