Please use this identifier to cite or link to this item: http://hdl.handle.net/10397/3787
Title: Energy efficiency in IEEE 802.11 wireless networks
Authors: Xie, Yi
Keywords: Hong Kong Polytechnic University -- Dissertations
Computer networks
Energy conservation
Issue Date: 2008
Publisher: The Hong Kong Polytechnic University
Abstract: To achieve energy efficiency, various existing mechanisms investigate the tradeoff between communication quality and energy consumption. This thesis proposes two new schemes based on Transmission Power Control (TPC) and Power Saving Mode (PSM). Both of them are Access Point (AP) centric and support multiple clients in an IEEE 802.11 infrastructure network. The new TPC-based scheme uses a polling-based MAC, such as the point coordination function (PCF) in IEEE 802.11, to optimize energy efficiency of all clients. The new PSM-scheme, on the other hand, achieves the same goal by using distribution coordination function (DCF). The problem of optimizing energy efficiency in a polling-based network is to determine optimal transmission power allocations (therefore the transmission rates) for all clients. The AP notifies the optimal transmission power to all clients at the beginning of each polling cycle. Using this optimal AP-centric TPC, the total energy consumed by all clients in a polling cycle will be minimized when their buffer queues are guaranteed stable. We have solved this problem by first formulating it as a stability-constrained optimization problem and then solving it using an iterative algorithm. The problem is tackled by using two polling schedules: Phase Grouping (PC) and Mobile Grouping (MG). Our extensive experiment results have shown that the optimal power allocations can improve the energy efficiency of a random power allocation by saving energy as much as four times. After using the AP-centric TPC, the optimal MG scheme is more energy efficient than the optimal PG scheme, because each client in the PG scheme spends more energy during the reception period. We have also shown that there are incentives for the clients to adopt the optimal power allocations. The problem of optimizing energy efficiency in a DCF network using PSM is to determine the optimal PSM parameters, e.g., beacon interval (BI), listen interval (LI) and minimal contention window (CW) for each client. The centralized PSM (C-PSM) deploys these optimal parameters on the AP and clients. The BI and LIs are optimized to reduce major energy consumption due to unnecessary wake-ups and channel contention. Both simultaneous wake-ups and infrequent wake-ups will cause channel contention during which the contending clients waste much of their energy in the idle mode. In addition, we have proposed a first-wake-up schedule to further reduce simultaneous wake-ups. The CWs, on the other hand, are tuned to balance the clients' access probabilities, such that a shorter backoff period is assigned to a client waking up with a lower frequency. Our extensive simulation results have shown that compared with the standard PSM C-PSM reduces power consumption by up to 76%, increases energy efficiency as high as 320% and shortens AP buffering delay as much as 88%. Moreover, the first-wake-up schedule saves the energy efficiency by up to 22%. C-PSM's advantage also increases with the number of clients in the network.
Description: viii, 213 leaves : ill. ; 30 cm.
PolyU Library Call No.: [THS] LG51 .H577P COMP 2008 Xie
URI: http://hdl.handle.net/10397/3787
Rights: All rights reserved.
Appears in Collections:Thesis

Files in This Item:
File Description SizeFormat 
b22397589_link.htmFor PolyU Users 162 BHTMLView/Open
b22397589_ir.pdfFor All Users (Non-printable) 3.52 MBAdobe PDFView/Open
Show full item record

Page view(s)

432
Last Week
2
Last month
Checked on Feb 12, 2017

Download(s)

338
Checked on Feb 12, 2017

Google ScholarTM

Check



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.