A Brain-Computer-Interface for Multi-Modal Mobile Device InputBachelor Thesis
8 March 2024, by Cedric Robert

Photo: midjourney
This thesis explains the design and development of a simple Brain-Computer Interface, which can be prepared in minutes and uses an Neural Network to classify the input for the Accessibility-API to interact with Android mobile devices.
The created prototype covers the entire workflow of creating a personalized Neural Network for the thought classification of each user, as well as live communication between an off-the-shelf gel-free EEG cap over a Raspberry Pi and an Android mobile device. An evaluation has shown that, with the current setup, this system is not usable in everyday life. None of the participants were able to enter input willingly. Furthermore, this thesis attempts to shed light on the possible reasons for this result and discusses ways to improve the system.
Supervised by:
Prof. Dr. Janick Edinger, Tim Rolff