TY - GEN
T1 - End-to-end rate selection for opportunistic reception in multi-rate wireless networks
AU - Kumar, Raju
AU - Eswaran, Sharanya
AU - Porta, Thomas La
PY - 2011
Y1 - 2011
N2 - In this paper we propose an end-to-end algorithm, called NUM-RS, for jointly selecting link transmission rates and source rates in a multi-hop multi-rate wireless network. Prior works on rate selection, including those that explicitly account for opportunistic reception, perform rate selection on a hop-by-hop basis, attempting to maximize the throughput on each link. Our algorithm leverages the Network Utility Maximization (NUM) framework, thus providing end-to-end semantics for rate selection and proportional fairness with low overhead. By using end-to-end semantics NUM-RS considers both source rates and congestion in the vicinity of links used by a flow when selecting link rates. Our results show that NUM-RS increasingly outperforms contemporary hop-by-hop rate selection schemes as the number of hops in the flows increase. For example, 20% and 50% of 8-hop flows exhibit performance gains of at least 36% and 15%, respectively, in terms of end-to-end throughput. In some cases, gains of up to 80% can be achieved.
AB - In this paper we propose an end-to-end algorithm, called NUM-RS, for jointly selecting link transmission rates and source rates in a multi-hop multi-rate wireless network. Prior works on rate selection, including those that explicitly account for opportunistic reception, perform rate selection on a hop-by-hop basis, attempting to maximize the throughput on each link. Our algorithm leverages the Network Utility Maximization (NUM) framework, thus providing end-to-end semantics for rate selection and proportional fairness with low overhead. By using end-to-end semantics NUM-RS considers both source rates and congestion in the vicinity of links used by a flow when selecting link rates. Our results show that NUM-RS increasingly outperforms contemporary hop-by-hop rate selection schemes as the number of hops in the flows increase. For example, 20% and 50% of 8-hop flows exhibit performance gains of at least 36% and 15%, respectively, in terms of end-to-end throughput. In some cases, gains of up to 80% can be achieved.
UR - http://www.scopus.com/inward/record.url?scp=80052790449&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=80052790449&partnerID=8YFLogxK
U2 - 10.1109/SAHCN.2011.5984948
DO - 10.1109/SAHCN.2011.5984948
M3 - Conference contribution
AN - SCOPUS:80052790449
SN - 9781457700934
T3 - 2011 8th Annual IEEE Communications Society Conference on Sensor, Mesh and Ad Hoc Communications and Networks, SECON 2011
SP - 601
EP - 609
BT - 2011 8th Annual IEEE Communications Society Conference on Sensor, Mesh and Ad Hoc Communications and Networks, SECON 2011
T2 - 2011 8th Annual IEEE Communications Society Conference on Sensor, Mesh and Ad Hoc Communications and Networks, SECON 2011
Y2 - 27 June 2011 through 30 June 2011
ER -