Acoustic monitoring of human activity using sensor network

OData support
Supervisor:
Dr. Orosz György
Department of Measurement and Information Systems

The wireless sensor networks (WSN) are very fast evolving sector, which possible applications are in industrial, medical and home environment. The motes of the sensor networks can sense many environmental factors.

In this thesis I would like to implement acoustic monitoring of human activity with WSN. It needs to be familiar with the devices, then build my own network, and using Time Difference of Arrival principle. With this principle can be estimate the position of the voice source. We need the synchronization of the nodes, the positions and knowing the acoustic spreading. The estimation needs large computing capacity, so the system will have a pc with Matlab. I would like to invest the working, the limitations and the possibility of improvement of the whole system.

As a part of the work I have got to know the Berkeley MICAz motes, the operation system, programming language, network topologies, the structure and the operation of flash memories, besides later I will meet and study synchronization methods, voice detection and recognition possibilities and the theory of the acoustic positioning, then the implementation.

Downloads

Please sign in to download the files of this thesis.