Electrical and Electronic Engineering - Theses

Permanent URI for this collection

Search Results

Now showing 1 - 1 of 1
  • Item
    Thumbnail Image
    Measurement in 802.11 wireless networks and its applications
    Ahmad Yar Khan, Malik ( 2010)
    Ease of deployment, wireless connectivity and ubiquitous mobile on-the-go computing has made the IEEE 802.11 the most widely deployed Wireless Local Area Network (WLAN) standard in the world. The wireless channel is fundamentally different to its wired counterpart and exhibit characteristics which are difficult to model. We therefore revert to measurement based characterization of wireless networks. A wireless network testbed was thus developed using off-the-shelf wireless cards. Available bandwidth measurement is particularly challenging in the wireless environment because of adaptive data rates, a time varying channel, and CSMA/CA based contention instead of simple FIFO queueing. We present and experimentally evaluate a novel available bandwidth estimation scheme, ‘SPEEDO’, for 802.11 networks in infrastructure mode, based on passive monitoring, without any need for access point cooperation or protocol modifications. An ability to accurately classify observed packet errors according to their root cause: physical layer or MAC layer contention, in 802.11 networks, opens up many opportunities for performance improvement at both, the MAC and IP layers. We investigate three approaches to isolate physical errors from contention ones based on: channel utilisation, error correlation and fragmentation based frame reservation. We implemented these approaches on our testbed and show that fragmentation technique outperforms the other approaches. We show that current rate adaptation algorithms in the IEEE 802.11 suffer under congested scenarios because of their inability to isolate the physical error from the contention. We introduce and compare two variants of a single core idea enabling the isolation and accurate measurement of physical packet error, based on exploiting existing features of the MAC standard in a novel way. One is based on the RTS/CTS mechanism, and the other on packet fragmentation. Using experimental results from a wireless testbed, we show these mechanisms can be used to improve the performance of two existing algorithms, SampleRate and AMRR, both for individual stations and for the system as a whole. Finally, we present ‘SmartRate’, a highly adaptive, throughput, congestion and environment aware rate adaptation algorithm. It is designed to avoid the weaknesses inherently present in the current rate adaptation algorithms and is efficient, robust and can readily adapt according to the situation (stationary or mobile). It is shown to outperform both, SampleRate and AMRR, in single and multi user (congested) scenarios. The theme of this dissertation can be summarized as the quest to understand, characterize and improve the behavior of wireless links by using real live measurements for the benefit of network users. The work presented contributes to the field of wireless networks by: development of an available bandwidth estimation technique, extraction of wireless link properties like physical PER, improvement of existing rate adaptation algorithms by incorporation of physical PER, and finally by the development of a more robust and dynamic rate adaptation algorithm.