Instantaneous ber estimation based link adaptation and scheduling in multi-band ofdm uwb wpans

Maşazade, Engin (2006) Instantaneous ber estimation based link adaptation and scheduling in multi-band ofdm uwb wpans. [Thesis]

[thumbnail of masazadeengin.pdf] PDF
masazadeengin.pdf

Download (3MB)

Abstract

Ultra Wide Band (UWB) systems have gained interest bcaus of their abilities to provide high data rate with low power transmission for WPANs. This thesis considers th Multi band OFDM UWB WPANs over time varying channels. Due to the channel impairments, the links are subject to degradations in quality varying with time and space. To maximiz the link efficiency, we consider an immediate acknowledgement protocol with dynamically and optimally selected frame sizes according to the channels bit error rate (BER) estimates. The BER is first estimated in maximum likelihood sense, by applying Modiano's algorithm. Since, this makes the tracking of the channel difficult under fast fading conditions, we propose a n w method to model and estimate the BER of the MB-OFDM UWB links based on the OFDM subcarrir signal to noise ratios. In a wireless network, opportunistic scheduling improves the network performance by allocating more resources to good links. We also propose an opportunistic channel access scheme using IEEE 802.15.3 WPAN superframe structure to enhance th system throughput further. In this method, time allocations for individual links are proportionally determined with respct to their instantaneous efficiencies. Simulation results show that opportunistic scheduling along with frame size adaptation improves the individual link throughputs up to thirty percent for low number of links and up to hundred and ten percent for large number of links; and overall network throughputs by twenty seven prcent up to eighten simultaneously operating connections as compared to the equal-time scheduled links operating with fixed frame sizes.
Item Type: Thesis
Uncontrolled Keywords: OFDM. -- Ultra wide band, UWB. -- IEEE 802.15.3 medium access. -- BER estimation. -- Time scheduling
Subjects: T Technology > TK Electrical engineering. Electronics Nuclear engineering
Divisions: Faculty of Engineering and Natural Sciences > Academic programs > Computer Science & Eng.
Faculty of Engineering and Natural Sciences
Depositing User: IC-Cataloging
Date Deposited: 15 Apr 2008 08:46
Last Modified: 26 Apr 2022 09:47
URI: https://research.sabanciuniv.edu/id/eprint/8371

Actions (login required)

View Item
View Item