Date of Award
Doctor of Philosophy (PhD)
This dissertation focuses on clustering and hybrid routing in Mobile Ad Hoc Networks (MANET). Specifically, we study two different network-layer virtual infrastructures proposed for MANET: the explicit cluster infrastructure and the implicit zone infrastructure. In the first part of the dissertation, we propose a novel clustering scheme based on a number of properties of diameter-2 graphs to provide a general-purpose virtual infrastructure for MANET. Compared to virtual infrastructures with central nodes, our virtual infrastructure is more symmetric and stable, but still light-weight. In our clustering scheme, cluster initialization naturally blends into cluster maintenance, showing the unity between these two operations. We call our algorithm tree-based since cluster merge and split operations are performed based on a spanning tree maintained at some specific nodes. Extensive simulation results have shown the effectiveness of our clustering scheme when compared to other schemes proposed in the literature. In the second part of the dissertation, we propose TZRP (Two-Zone Routing Protocol) as a hybrid routing framework that can balance the tradeoffs between pure proactive, fuzzy proactive, and reactive routing approaches more effectively in a wide range of network conditions. In TZRP, each node maintains two zones: a Crisp Zone for proactive routing and efficient bordercasting, and a Fuzzy Zone for heuristic routing using imprecise locality information. The perimeter of the Crisp Zone is the boundary between pure proactive routing and fuzzy proactive routing, and the perimeter of the Fuzzy Zone is the boundary between proactive routing and reactive routing. By adjusting the sizes of these two zones, a reduced total routing control overhead can be achieved.
"Clustering and Hybrid Routing in Mobile Ad Hoc Networks"
(2005). Doctor of Philosophy (PhD), Dissertation, Computer Science, Old Dominion University, DOI: 10.25777/f5th-ns52