Energy-efficient sleep interval decision scheme for IEEE 802.11 wireless LAN access points

Abstract

IEEE 802.11 wireless local area networks have been widely deployed in public and private areas in recent years. Many WLAN devices have been installed in many types of electronic devices such as personal computers, smartphones, etc. and energy consumption in wireless networks is expected to account for almost 50\% in information and communication technology in 2020. It also affects operational expenditure of network service providers. The IEEE 802.11 standards only support power saving control for stations which mostly have battery limitations. It does not support power saving control for AP because it is assumed that APs are always supplied by main electric power. However, to support increasing demands of wireless local area network, the number of wireless access points are increased. On the other hands, traffic profiles indicates that access points have many chance to reduce its energy consumption due to low traffic load. To save energy, energy efficient sleep mechanism for access point is needed. In this thesis, energy-efficient sleep interval decision scheme for IEEE 802.11 wireless LAN access points is proposed. Proposed scheme uses dynamic sleep boundary decision algorithm which dynamically set upper and lower bound of sleep duration considering traffic arrival rate and traffic delay requirements. By analyzing bulk arrival queueing model, numerical results show proposed scheme can increase energy efficiency about 25\% with reasonable increment of packet delay in off-peak situation.