Energy Efficiency in Wireless Access Networks: Measurements, Models and Algorithms

Wireless Telecommunication networks have become fundamental to daily activities. Today, people have access to at least one type of wireless telecommunication network. In this context, optimizing the energy consumption of wireless telecommunications infrastructure has become a new challenge for the research community, governments and industries in order to reduce CO2 emission and operational energy costs. This thesis investigates the power consumption of indoor/outdoor Wireless Access Devices (WADs, specifically WiFi and WiMAX access points) and provides novel techniques for improving the energy efficiency of wireless access networks. Our approach focuses on monitoring and analyzing the power consumption of WADs using real-testbed and experimental measurements in order to understand the fundamental limits and trade-offs involved. This, in turn, will be used to propose efficient techniques to reduce power consumption and to maximize the energy efficiency of wireless access networks. We introduce energino a novel hardware and software solution for real-time energy consumption monitoring in wireless networks. We also propose an experimentally-driven approach to (i) characterize typical WADs from a power consumption standpoint, (ii) develop simple and accurate power consumption models and metrics for such WADs, and (iii) design techniques to tune the power consumption of a wireless infrastructure to the actual network conditions in terms of both users density and traffic patterns. Our measurements from several real-life deployments show that (a) the power consumption of such WADs exhibits a linear dependence on the traffic until a saturation point is reached and (b) the developed techniques can deliver significant energy savings with minimal degradation in terms of the quality of service provided.