Optimization of User Association for Wireless Local Area Networks

The Hong Kong University of Science and Technology
Department of Computer Science and Engineering


PhD Thesis Defence


Title: "Optimization of User Association for Wireless Local Area Networks"

By

Mr. Wang Kit WONG


Abstract

IEEE 802.11 wireless local area networks (WLANs) have been widely deployed to 
provide ubiquitous broadband Internet access for wireless users. Users are 
often covered by multiple access points (APs) in today’s WLAN deployment. User 
association control is to assign or associate users in the overlapping regions 
to one of the APs for service. The default implementation always associates a 
user to the best signal AP. This leads to unsatisfactory throughput performance 
at congested APs and channel under-utilization at others. The performance of 
WLANs is highly related to user association.

We first study the joint user association and random access control 
optimization problem. Random access control is to determine APs’ contention 
windows and hence transmit probabilities, which is affected by users associated 
with it. We propose a distributed algorithm termed CARA (Joint Client 
Association and Random Access Control) to tackle the problem, which iteratively 
optimizes user association and random access control with optimality guarantee. 
Through extensive simulation and experiment, we show that CARA achieves 
substantially higher throughput than the state-of-the-art schemes.

Secondly, we generalize our study to WLANs with coordinated beamforming (CB) 
capabilities. CB has emerged as a promising interference coordination approach 
for co-operative multiple-input-multiple-output (MIMO) systems. CB allows 
interfering APs to transmit as a group. Due to heavy channel state information 
(CSI) feedback overhead, APs need to be partitioned into cooperation groups no 
larger than a certain size where only APs in the same group are able to 
cooperate with CB. We study the novel optimization problem of minimizing AP 
load by joint AP grouping and user association as they are coupled. Based on 
alternating direction optimization, we propose DAGA (Distributed Joint AP 
Grouping and User Association) to tackle the optimization problem. DAGA 
produces an approximated user association solution which is at most elogm (m is 
the number of APs) times of the optimum.

Finally, we consider user migration (re-association) cost. To effect user 
association optimization, a user needs to handshake with the target AP for 
authentication and association. Therefore, there is a certain migration cost, 
because of the traffic and management overhead. We study the user association 
problem for WLANs with migration cost constraint and propose a factor-4 
approximation algorithm to tackle the problem, which achieves similar 
performance as previous schemes but incurs much less user migration cost.


Date:			Thursday, 27 June 2019

Time:			2:30pm - 4:30pm

Venue:			Room 3494
 			Lifts 25/26

Chairman:		Prof. Wing-Hung Ki (ECE)

Committee Members:	Prof. Gary Chan (Supervisor)
 			Prof. Cunsheng Ding
 			Prof. Yangqiu Song
 			Prof. Roger Cheng (ECE)
 			Prof. Yiu-Bun Lee (CUHK)


**** ALL are Welcome ****