print page

Project Work

Evaluation and Comparison of Prediction Models for a Solar Energy Harvester

In the Institute of Telematics, an energy harvesting power supply with a supercapacitor as its energy buffer and a solar cell as its power source has been developed. Due to the novelty of the device, no models and estimations of incoming energy are available yet. However, they are required in order to allow for valid and fine-grained lifetime prediction of sensor nodes. Once the amount and times of incoming energy can be estimated, the availability of future energy resources become predictable. This in turn enables energy-aware scheduling in order to circumvent energy starvation, which would lead to lost or corrupted sensor readings.

The goal of this thesis is to develop a model for the amount of energy gathered by the solar cell of the harvester module depending on the environment and weather conditions. Furthermore, this model has to be validated and shall be transformed into an algorithm for real sensor node hardware, so that nodes are enabled to estimate the time and amount of incoming energy, finally yielding the possibility to realize energy-aware scheduling.

Start date 4. January 2010
End date 30. April 2010
Documents Flyer
Projects EBS | SensorNet
Supervisor Dr.-Ing. Christian Renner