Maintaining Performance while Saving Energy on Wireless LANs

Minimizing the energy consumption of wireless network interfaces is crucial in the design of battery-powered mobile computing devices. The IEEE 802.11 wireless LAN specification describes a power-saving mode which allows the network interface on the mobile device to enter a sleep mode when the link is idle; periodically, the mobile device must wake up to listen to beacons from the access point which indicate if it has any buffered data. This work first presents detailed measurements which show that when TCP is run over existing implementations of the 802.11 power-saving mode, performance suffers because otherwise fast round-trip-times are rounded up to 100 ms. The proposed solution is the Stay-Awake scheme which delays entering sleep mode for a short period of time after the link is active. This work also determines that in typical web browsing traffic scenarios existing 802.11 power-saving mode implementations will spend most of their energy sleeping and listening to beacons during relatively long idle periods. The proposed improvement is ListenInterval-Backoff which allows the network interface to sleep for longer periods of time when the link remains idle. The combination of these proposed power-saving enhancements is shown to reduce the additional per-page delay incurred by the power-saving mode from a mean of around 70% to 3% with negligible change in the overall energy consumption, or to reduce the per-page delay to 10% while reducing the overall energy by 20%.

[1]  Krishna M. Sivalingam,et al.  Performance comparison of battery power consumption in wireless multiple access protocols , 1999, Wirel. Networks.

[2]  Bruce A. Mah,et al.  An empirical model of HTTP network traffic , 1997, Proceedings of INFOCOM '97.

[3]  Krishna M. Sivalingam,et al.  A Survey of Energy Efficient Network Protocols for Wireless Networks , 2001, Wirel. Networks.

[4]  Suresh Singh,et al.  PAMAS—power aware multi-access protocol with signalling for ad hoc networks , 1998, CCRV.

[5]  Adam Wolisz,et al.  Power-saving mechanisms in emerging standards for wireless LANs: the MAC level perspective , 1998, IEEE Wirel. Commun..

[6]  Nathan Ickes,et al.  Physical layer driven algorithm and protocol design for energy-efficient wireless sensor networks , 2002, MobiCom 2001.

[7]  Surendar Chandra,et al.  Wireless network interface energy consumption implications of popular streaming formats , 2001, IS&T/SPIE Electronic Imaging.

[8]  Robert Tappan Morris,et al.  Span: An Energy-Efficient Coordination Algorithm for Topology Maintenance in Ad Hoc Wireless Networks , 2002, Wirel. Networks.

[9]  Eric A. Brewer,et al.  System Design Issues for Internet Middleware Services: Deductions from a Large Client Trace , 1997, USENIX Symposium on Internet Technologies and Systems.

[10]  Robin Kravets,et al.  Application‐driven power management for mobile communication , 2000, Wirel. Networks.

[11]  Fred Douglis,et al.  Adaptive Disk Spin-Down Policies for Mobile Computers , 1995, Comput. Syst..

[12]  Martin Nilsson,et al.  Investigating the energy consumption of a wireless network interface in an ad hoc networking environment , 2001, Proceedings IEEE INFOCOM 2001. Conference on Computer Communications. Twentieth Annual Joint Conference of the IEEE Computer and Communications Society (Cat. No.01CH37213).

[13]  Paul Horton,et al.  A Quantitative Analysis of Disk Drive Power Management in Portable Computers , 1994, USENIX Winter.