Applied data mining techniques in sensor networks

OData support
Supervisor:
Dr. Szabó Sándor
Department of Networked Systems and Services

According to several statistics and predictions the 60 million Smart Meter devices used in the EU will be doubled in their numbers during the next two years, and this outstanding progress will most likely have some impact on Hungary too. The change in the demand for intelligent sensor networks will be even greater, since they have several notable advantages, particularly when the information gathered out of the metered data is being put to use efficiently, supported by an information system.

The aim of my thesis is to design and construct the basic data processing of an intelligent sensor network, its database architecture, and a user interface with various alarm functions included.

In my thesis I will present the RemoteSense intelligent sensor network in details, for which I will provide a personal data gathering solution for its problem with data processing, using the analysis of existing networks and systems, and their alarming functions. My user interface will be able to recognize various anomalies - for example a pipe breakage – relying on the incoming data, furthermore with suitable settings it will be able to detect even the abusing attempts made against the network.

My designed system in its current form and with its current settings can be used in a more universal way, for example it is able to measure the water, electricity and gas consumption, hence it can be used in a much wider scale, with a few modifications it can support more specified and task oriented environments too.

Downloads

Please sign in to download the files of this thesis.