Hot Motion detection in RFID-enabled wireless sensor networks
(project, diploma, bachelor's, master's thesis)
Wireless sensor networks are becoming increasingly popular in a variety of monitoring and surveillance applications. Simultaneously, Radio Frequency IDentification (RFID) technology has been tremendously fostered in real object tracking applications including supply chain and asset tracking scenarios. Lately, several approaches have been taken towards integrating the aforementioned technologies in real applications.
This project focuses on traffic data analysis, in a RFID-enabled wireless sensor network. The network is organized in a 3-level hierarchy as it is shown in the figure below. Each leader covers a particular region of the network and is aware only of the activity that takes part in this part of the network.
We are interested in real time hot motion path extraction in such a network. Hot motion paths are popular routes within the road network, that is, routes preferred by many objects moving on it. The goal of this project is to study methods for hot motion path detection. We distinguish between:
(i) the centralized case, where all data are propagated into a central server where they are analyzed and
(ii) the de-centralized case where the data are analyzed locally at each leader node and only summary information is propagated to the central server.
For the extraction of hot motion paths, we will employ frequent itemsets mining.
Requirements
- KDD/ Data Mining concepts (KDD I/ KDD II)
- Good programming skills (Java)
- Independent work
Contact
If you are interested in this topic and/or if you have further questions please contact:Eirini Ntoutsi
(This is joint project with
Nikos Giatrakos from University of Piraeus, Grece)
