Load Aware Channel-Width Assignments in Wireless LANs

Measurements studies show that there exist large spatial and temporal fluctuations in the traffic load handled by different access points in Wireless LANs. In order to alleviate this problem, researchers have proposed various load-balancing techniques based for instance on channel assignment, power control, or client allocation. Fundamentally, however, assigning each AP the same amount of bandwidth (one channel) can inevitably lead to inefficient usage of the spectrum. In this work, we address the problem by adaptively tuning a radio parameter that has so far been largely untouched in Wireless LAN networks: the channel-width. Particularly, we show that a significant improvement in network capacity and per-client fairness can be achieved if the channel-widths at different APs are made a function of the traffic load. We propose the use of dynamic-width channels, where every AP adjusts its center-frequency and channel width to match its current traffic load. Our techniques are made possible by recent advances in radio hardware design and do not require changes in current hardware. We demonstrate the effectiveness of our scheme through analysis and simulations using real-world scenarios.

[1]  Alec Wolman,et al.  A Location-Based Management System for Enterprise Wireless LANs , 2007, NSDI.

[2]  Yunnan Wu,et al.  Allocating dynamic time-spectrum blocks in cognitive radio networks , 2007, MobiHoc '07.

[3]  Thomas M. Cover,et al.  Elements of Information Theory , 2005 .

[4]  Roger Wattenhofer,et al.  Protocol Design Beyond Graph-Based Models , 2006, HotNets.

[5]  Konstantina Papagiannaki,et al.  The need for cross-layer information in access point selection algorithms , 2006, IMC '06.

[6]  Paramvir Bahl,et al.  Cell Breathing in Wireless LANs: Algorithms and Evaluation , 2007, IEEE Transactions on Mobile Computing.

[7]  Konstantina Papagiannaki,et al.  Optimal design of high density 802.11 WLANs , 2006, CoNEXT '06.

[8]  John G. Proakis,et al.  Digital Communications , 1983 .

[9]  Magdalena Balazinska,et al.  Characterizing mobility and network usage in a corporate wireless local-area network , 2003, MobiSys '03.

[10]  A. Gräf,et al.  On Coloring Unit Disk Graphs , 1998, Algorithmica.

[11]  L BeckLeland,et al.  Smallest-last ordering and clustering and graph coloring algorithms , 1983 .

[12]  Aravind Srinivasan,et al.  Client-driven channel management for wireless LANs , 2006, MOCO.

[13]  Mary Baker,et al.  Analysis of a local-area wireless network , 2000, MobiCom '00.

[14]  Yunnan Wu,et al.  KNOWS: Kognitiv Networking Over White Spaces , 2007 .

[15]  Lili Qiu,et al.  Estimation of link interference in static multi-hop wireless networks , 2005, IMC '05.

[16]  Srinivasan Keshav,et al.  SMARTA: a self-managing architecture for thin access points , 2006, CoNEXT '06.

[17]  David Kotz,et al.  Analysis of a Campus-Wide Wireless Network , 2002, MobiCom '02.

[18]  Lili Qiu,et al.  Impact of Interference on Multi-Hop Wireless Network Performance , 2003, MobiCom '03.

[19]  Jochen Könemann,et al.  Faster and simpler algorithms for multicommodity flow and other fractional packing problems , 1998, Proceedings 39th Annual Symposium on Foundations of Computer Science (Cat. No.98CB36280).

[20]  Ravi Jain,et al.  Model T: an empirical model for user registration patterns in a campus wireless LAN , 2005, MobiCom '05.

[21]  William A. Arbaugh,et al.  Partially overlapped channels not considered harmful , 2006, SIGMETRICS '06/Performance '06.