Please use this identifier to cite or link to this item:
Title: Data hovering algorithm for improving data retention and data quality in energy-constrained mobile wireless sensor networks
Authors: He, Yingjie
Issue Date: 2012
Publisher: Newcastle University
Abstract: A Wireless Sensor Network (WSN) is composed of numerous spatially distributed, low cost, low power and multifunctional sensor nodes which can be used to monitor the surrounding environment. In mobile networks, the sensed data collected by the sensor nodes may move out of the area where it has been gathered (area of origin) with its carrying node. A problem may arise in this situation: when requesting the historical information of a specific area, it is possible that none of the nodes currently located in such area can provide the required information. This thesis addresses the issue of retaining data it its area of origin in an energy-constrained, infrastructure-less mobile Wireless Sensor Network. The concept of this “Data Hovering” has been defined in which the location-based data hovers in its area of origin by transmission between network nodes. Based on this concept, several policies need to be defined as well as considering the constraints of WSN including limited energy and limited transmission bandwidth. The existing related work has then been investigated by examining how it proposed to define the Data Hovering policies, in order to explore the limitations. It has been found that the existing approaches are not well suited to mobile WSN, due to the unique characteristics of WSN. In this thesis, an autonomous Data Hovering algorithm consisting of defined policies has been designed to improve the data retention (data availability) and the quality of the retained data which ensures that the retained data represents different information. The defined Data Hovering algorithm has been implemented in a network simulator and a baseline with simple policies has also been selected in order to be compared with the defined policies. The evaluation in terms of data availability, data quality and energy consumption has then been carried out to analyze the behaviours of the defined algorithm. Finally, the potential future work has been suggested.
Description: PhD Thesis
Appears in Collections:School of Computing Science

Files in This Item:
File Description SizeFormat 
He 12.pdfThesis1.64 MBAdobe PDFView/Open
dspacelicence.pdfLicence43.82 kBAdobe PDFView/Open

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.