On supporting power-efficient streaming applications in wireless environments

Reducing the power consumption of the wireless network interface (WNl) is an effective way to prolong the battery lifetime of the mobile terminal. It takes some time for the WNI to transit from the power-saving mode to the active mode. This transition delay and the error-prone wireless link bring many challenges for designing power-aware and QoS-aware service models. In this paper, we present a novel power-conserving service model for streaming applications over wireless networks. At the base station side, a new scheduling algorithm, called rate-based bulk scheduling (RBS), is designed to decide which flow should be served at which time. The mobile terminal relies on a proxy to buffer data so that the WNI can sleep for a long time period to save power. To deal with channel errors, a novel adaptive technique is presented to adjust the sleep time of the WNI according to the channel condition. Through analysis, we prove that RBS can provide delay guarantee and it is more power efficient than other rate-based fair queuing algorithms. We use audio-on-demand as a case study to evaluate the performance of RBS. Experimental results show that RBS achieves excellent QoS provision for each flow and significantly reduces the power consumption.