Information-Driven Mobility Control in Mobile Sensor Networks
thesisposted on 21.07.2015 by Wen Jiang
In order to distinguish essays and pre-prints from academic theses, we have a separate category. These are often much longer text based documents than a paper.
A mobile sensor network is a powerful tool used to monitor physical phenomena and to provide services over an area of interest. This dissertation addresses mobility control problems, which occur when a mobile sensor network is used to perform multiple simultaneous tasks in addition to coverage. First, we provide a distributed iterative motion control algorithm through cubic and bicubic spline interpolation for scalar field estimation. We prove the convergence of the algorithm that solves for the spline coefficients. We incorporate this algorithm into a coverage control problem whereby we estimate event occurrence density function that drives the mobile sensor nodes to an optimal sensing configuration. Second, we model the cost of information aggregation and develop a distributed motion control algorithm for the network to achieve optimal configuration under three basic types of network structures. In order to ensure low data query latency, the hierarchical network structure has to be well-balanced. We formulate the problem of a query-aware constrained coverage with information aggregation. Within this formalism, we study how routing constraints can be enforced in coverage control with information aggregation. Consequently, we develop a motion control algorithm that drives the mobile sensor nodes to an optimal configuration while ensuring low latency of query processing. Third, we generalize information aggregation to information dissemination where information generated in the area is disseminated to multiple or even an infinite number of destinations in the same area. We study how information dissemination can be modeled through optimization of an appropriate cost function and propose a distributed motion control algorithm that drives the mobile sensor nodes to an optimal configuration.