This dissertation focuses on the design and analysis of distributed primitives for
mobile ad hoc networks, in which mobile hosts are free to move arbitrarily. Arbitrary
mobility adds unpredictability to the topology changes experienced by the network, which
poses a serious challenge for the design and analysis of reliable protocols. In this work,
three different approaches are used to handle mobility. The first part of the dissertation
employs the simple technique of ignoring the mobility and showing a lower bound for the
static case, which also holds in the mobile case. In particular, a lower bound on the worstcase
running time of a previously known token circulation algorithm is proved. In the
second part of the dissertation, a self-stabilizing mutual exclusion algorithm is proposed
for mobile ad hoc networks, which is based on dynamic virtual rings formed by circulating
tokens. The difficulties resulting from mobility are dealt with in the analysis by showing
which properties hold for several kinds of mobile behavior; in particular, it is shown that
mutual exclusion always holds and different levels of progress hold depending on how
the mobility affects the token circulation. The third part of the dissertation presents two
broadcasting protocols which propagate a message from a source node to all of the nodes in
the network. Instead of relying on the frequently changing topology, the protocols depend
on a less frequently changing and more stable characteristic â the distribution of mobile
hosts. Constraints on distribution and mobility of mobile nodes are given which guarantee
that all the nodes receive the broadcast data.
Identifer | oai:union.ndltd.org:tamu.edu/oai:repository.tamu.edu:1969.1/4294 |
Date | 30 October 2006 |
Creators | Chen, Yu |
Contributors | Welch, Jennifer L. |
Publisher | Texas A&M University |
Source Sets | Texas A and M University |
Language | en_US |
Detected Language | English |
Type | Book, Thesis, Electronic Dissertation, text |
Format | 714595 bytes, electronic, application/pdf, born digital |
Page generated in 0.0042 seconds