The objective of the project SmartSleep is to analyze a huge source of biometric data, to detect patterns and to derive relationships between a person’s behavior and sleep disorders. In order to capture the data, mobile sensors connected 24 hours per day will be used. Besides the movement, these sensors will detect additional parameters as input to clinical long-term studies. One task is to provide mechanisms pre-processing the data stream before pushing it to the backbone servers. So, even the patient may receive feedback in real-time. A second task is to provide data analysis components beyond current state of the art systems. The goal is to close the gap in existing data-mining algorithms by extending them with scenarios relevant for sleep disorders. The overall project result will provide a remarkable advancement over a purely statistical relationship analysis within a population and it offers a perspective to achieve a system efficiently customizable to individual needs of a patient without strong intervention.