Model Based Data Compression for an ECG Signal (Bachelorthesis)


Claus Philipp Koppermann

31.10.2013, 10:00, room 4981


This bachelor’s thesis addresses the development of a model based data compression regarding an ECG Signal. The main priority is to reduce the energy consumption in wireless sensor networks. A general overview of sensor nodes, body area network, wireless sensor networks and smart environment is imparted. Additionally the processing of an ECG with digital filters and common compression methods such as DPCM are discussed. The development of a model based data compression is performed with a real measured ECG signal and a simplified version, produced by a signal generator. Furthermore the new model based compression is compared with other simple compression methods. The final algorithm is described in its functionality and adapted on a microcontroller. Results show a stable compression method with a compression ratio relative to noise and sampling rate of an ECG signal.