Power management of a computer display

Abstract: The goal with this project is to investigate how the energy management can be improved on a linux computer. Focus lies on the screen back light as this is one of the major energy consumers in a laptop computer. In the systems available today the regulation is static and the back light is set in fixed steps without the user behavior in consideration. Adaboost is a machine learning algorithm that is implemented in a software daemon to regulate the back light adaptive. The algorithm is trained with data containing the user demands in the specific system state. Data covering key presses per time, actual brightness,battery level and the information if power cord is plugged in or not. The algorithm will continuously classify the current system state and increase or decrease the back light. A statistical study shows that the daemon consumes more energy than the original system. Future developments and improvements are discussed.