NeTS: Small: Power-Performance Tradeoffs for Mobile Devices in Next Generation WiFi Networks

点击翻译按钮获取中文摘要

With the tremendous growth in smartphones and other mobile devices, and the increasingly complex applications they run, the impact of the energy consumption is felt by every user. Smartphone batteries need to be charged too often or the devices are run in modes that limit their capabilities, to save energy. A major fraction of the energy consumption in smartphones comes from the WiFi radio. The next generation WiFi radios include a rich set of features offering increased bit rates, longer range, and more reliable coverage. However, these benefits often come at the cost of significantly higher power consumption. Consequently, smartphone manufacturers often restrict the features available to the users, thus trading performance for power savings. Also, the increasing demand for computational features such as multimedia in mobile devices has led to the use of complex computational hardware and peripherals for graphics and memory with higher contribution to the overall energy consumption. Hence, there is a need to develop a holistic approach to power analysis that considers both the communication and the computational paradigms, and novel protocols that efficiently utilize the battery power without compromising performance, which this projects aims to achieve. This work can have a lasting impact on better energy utilization in the mobile environment.This project seeks to develop models that capture the power-performance tradeoffs in WiFi-capable mobile devices and the impact of different features of the next generation WiFi standards on these tradeoffs. The models are to be developed based on extensive measurement studies using different mobile devices in a variety of scenarios. Further, it seeks to build realistic power models that consider diverse components of the system and their interactions, beyond just the CPU, as well as models for power-performance tradeoffs at the system level. Power analysis of the computational elements and peripherals will be developed, with considerations to varying factors that affect the power consumption, such as the applications, environment, and usage profile, through power modeling and measurements. The results will then be integrated with the WiFi models resulting in a holistic power model and profile. The developed models will guide the design of novel power management, rate adaptation, and WLAN management protocols that will allow mobile devices to realize more of their potential, while minimizing the power consumption and the impact to other WLAN clients. The project will bring increased awareness of energy consumption issues in modern mobile devices through outreach activities.