Scalable ad hoc and sensor networks

点击翻译按钮获取中文摘要

While technology cannot prevent disasters such as earthquake or tsunami, mobile ad hoc networks (MANETs) promise significant benefit to forecasting these and to the disaster relief afterwards.  A MANET is a wireless communication network that does not rely on existing infrastucture and supports mobile users.  Since nodes in such a network are not required to be able to communicate with each other directly, a MANET is usually multi-hop and intermediate nodes are often needed to relay messages between nodes in the network that are far apart.  In addition to applications such as disaster relief and rescue tasks, wireless sensor networks (WSNs) is one of the most implementations of the mobile ad hoc networking technology.  Despite many research activities in this area, the most significant factor hindering wider deployment of large MANETs is the scalability issue.  That is, the percentage of network bandwidth and device energy consumed by user data, and the per-node throughput decrease dramatically and approach zero as the number of nodes in the network, the node density, and movements of node increase.  To address the scalability issue, the approach employed in this project is to introduce a hierarchy among the network nodes to help overcome these problems, whether using device heterogeneity or superimposing logical structures.  This involves constructing a hierarchy with desired properties in a localized distributed fashion and maintaining it when the network conditions change.  The goal is to have a hierarchical structure that is easily maintained and will allow network protocols to scale well without consuming too much of the limited network resources.       To solve these problems, this project will resort to both analytical and experimental methods.  In particular, mathematical optimization and approximation will be used to formulate and solve these performance optimization problems.  On the other hand, to overcome the possible inaccuracy of analytical models, computer simulations will be used to verify the protocol performance.  More realistic and faithful performance evaluations will be carried out by the use of testbed networks that execute various protocols under study.