Electrocardiogram (ECG) is a noninvasive technology widely used in health care systems for diagnosis of heart diseases, and a wearable ECG sensor with long-term monitoring is necessary for real-time heart disease detection. However, the conventional ECG is restricted considering the physical size and power consumption of the system. In this thesis, we propose a Wireless ECG System with Bluetooth Low Energy (BLE) and Compressed Sensing (CS). The proposed Wireless ECG System includes an ECG sensor board based on a BLE chip, an Android application and a web service with a database. The ECG signal is first collected by the ECG Sensor Board and then transmitted to the Android application through BLE protocol. At last, the ECG signal is uploaded to the cloud database from the Android app. We also introduce Compressed Sensing into our system with a novel sparse sensing matrix, data compression and a modified Compressive Sampling Matching Pursuit (CoSaMP) reconstruction algorithm. Experiment results show that the amount of data transmitted is reduced by about 57% compared to not using Compressed Sensing, and reconstruction time is 64% less than using Orthogonal Matching Pursuit (OMP) or Iterative Re-weighted Least Squares (IRLS) algorithm. / Graduate

No description available.