Data aggregator for embedded devices

Paw Petersen

AbstractThis project deals with the creation a system which can act as a data aggregator for embedded devices. Having multiple embedded devices from which data can be retrieved using their own specic interface and protocol, the creation of the data aggregator would provide a nice unied interface for collecting data from these devices. Such embedded devices often use simple communication protocols, which are interfaced by the aggregator. The data from these devices will be collected, stored in a suitable database and presented in a user friendly way.
The aggregator is implemented using a single-board computer, the BeagleBone, which is used as a platform for running an embedded Linux distribution. The Linux distribution will be customized and compiled using a popular framework for building embedded Linux distributions.
To make the aggregator remotely upgradeable, a firmware update service is implemented. The firmware upgrade service makes use of several system partitions as well as cryptography to make the service resistant to system corruption due to power failures or wrong firmware files.
The system is developed to serve as a generic framework for instantiating data aggregators to be customized to work with various external devices. The project will feature an implementation of a data aggregator customized to work with a specific external device, a sensor, which will serve as an example of usage. The specific implementation is tested in a real setup using a virtual sensor to measure the performance of the aggregator.
The project concludes that a working data aggregator has been developed and tested. To use the data aggregator in a real scenario, there are still a couple of bugs which need to be fixed. Ideas for future development are listed in the report as well, the implementation of these should also be considered. The overall impression throughout this project is that the use of small computers like the BeagleBone seems to be a viable choice for creating devices like the data aggregator.
TypeBachelor of Engineering thesis [Academic thesis]
Year2013
PublisherTechnical University of Denmark, Department of Applied Mathematics and Computer Science
AddressMatematiktorvet, Building 303B, DK-2800 Kgs. Lyngby, Denmark, compute@compute.dtu.dk
SeriesB.Eng.-2013-39
NoteDTU supervisor: Sven Karlsson, svea@dtu.dk, DTU Compute
Publication linkhttp://www.compute.dtu.dk/English.aspx
BibTeX data [bibtex]
IMM Group(s)Computer Science & Engineering