Rate Adaptation (RA) for 802.11 has been deeply investigated in the past, in particular with the aim of achieving optimal RA with respect not only to channel-related errors but also to contention-related issues (i.e., collisions and variations in medium access times). Most of prior work in this field considered only RA from the point of view of a single node, i.e., evaluating the performance of different RA strategies adopted by the considered node in scenarios where other nodes use a fixed rate setting. In this paper, we analyze from a Game Theoretic perspective the case in which all users simultaneously perform RA. We show that state of the art strategies such as Goodput Optimal Rate Adaptation (GORA), in which every user selfishly tries to maximize his own performance accounting for issues such as collisions and medium access times, actually often results in degraded performance for all users, whereas simpler SNR-based RA schemes, which have been long regarded as sub-optimal, are actually much more robust.
A Game Theoretic evaluation of Rate Adaptation strategies for IEEE 802.11 based Wireless LANs
ZANELLA, ANDREA;
2009
Abstract
Rate Adaptation (RA) for 802.11 has been deeply investigated in the past, in particular with the aim of achieving optimal RA with respect not only to channel-related errors but also to contention-related issues (i.e., collisions and variations in medium access times). Most of prior work in this field considered only RA from the point of view of a single node, i.e., evaluating the performance of different RA strategies adopted by the considered node in scenarios where other nodes use a fixed rate setting. In this paper, we analyze from a Game Theoretic perspective the case in which all users simultaneously perform RA. We show that state of the art strategies such as Goodput Optimal Rate Adaptation (GORA), in which every user selfishly tries to maximize his own performance accounting for issues such as collisions and medium access times, actually often results in degraded performance for all users, whereas simpler SNR-based RA schemes, which have been long regarded as sub-optimal, are actually much more robust.Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.