Correlation based Rate Adaptation via Insights from Incomplete Observations in 802.11 Networks

Abstract—Rate adaptation, as a challenging issue for wireless purpose. Relatively little attention has been paid on identifying network design, has been an active research topic for years. whether root cause of transmission failure is correlated with Existing schemes either assume perfect channel information, or data rate. Such black-box based schemes usually lead to conduct rate adaptation in a black box way, hence can not throughput fluctuations and degradation. This observation achieve desirable performance. In this paper, we propose a novel motivates us to explore rate adjustment problem from scheme called Correlation based Rate Adaptation (CORA) to correlation perspective. address the rate adjustment problem. Unlike previous schemes, CORA splits rate into more atomic components and adjusts them On the other hand, nascent 802.11n [1] expands rate according to the correlation between rate adaptation actions and dependent parameters into spatial dimension by incorporating transmission results. We use IEEE 802.11n as the context for Multiple-Input-Multiple-Output (MIMO) technology. As a CORA design, where transmission mode has been expanded to result, different combinations of these parameters may mirror spatial dimension in addition to the usual modulation and the same data rate. Although with the same rate, these convolution coding mechanisms. Performance evaluation shows combinations behave quite differently under some similar that CORA can conduct rate adaptation in a more logical way channel conditions. Therefore, rate adaptation should be and significantly outperform the comparison scheme. conducted in a more atomic way