DocumentCode
2881522
Title
A Load-Balanced Route Selection for Network Coding in Wireless Mesh Networks
Author
Fan, Kai ; Wei, Xi ; Long, Dongyang
Author_Institution
Dept. of Comput. Sci., Sun Yat-sen Univ., Guangzhou, China
fYear
2009
fDate
14-18 June 2009
Firstpage
1
Lastpage
6
Abstract
COPE appears as the first wireless network coding approach which can improve the throughput of Wireless Mesh Networks (WMNs). But the established routes in COPE may result in fairly limited coding opportunities and low throughput gain. Although excessive intersections of flows can bring forth more coding opportunities, they may make some nodes overloaded, and decrease the throughput. Furthermore, detecting coding opportunities may induce the high routing delay. To solve these problems, we propose HLCR, a heuristic load-balanced coding-aware routing mechanism, and its novel path metric called Heuristic Path Metric for Coding and Load-balancing (HPMCL) in WMNs. HPMCL evaluates a path considering network load, expected number of transmissions and potential coding opportunities. Thus HLCR can efficiently find high throughput paths having coding opportunities and achieve good load balance with low cost by the HPMCL metric and other effective measures. Our simulation results in ns-2 show that HLCR scheme can gain higher network throughput compared with COPE.
Keywords
encoding; radio networks; resource allocation; telecommunication network routing; COPE; heuristic load-balanced coding-aware routing; heuristic path metric; load-balanced route selection; wireless mesh networks; wireless network coding; Broadcasting; Communications Society; Computer science; Delay; Network coding; Peer to peer computing; Routing protocols; Sun; Throughput; Wireless mesh networks;
fLanguage
English
Publisher
ieee
Conference_Titel
Communications, 2009. ICC '09. IEEE International Conference on
Conference_Location
Dresden
ISSN
1938-1883
Print_ISBN
978-1-4244-3435-0
Electronic_ISBN
1938-1883
Type
conf
DOI
10.1109/ICC.2009.5198624
Filename
5198624
Link To Document