Direct Analog-to-QRS detection front-end architecture for wearable ECG applications

The aim of this paper is to demonstrate a highly integrated architecture for direct Analog-to-QRS detection for ultra low power single chip implementation suitable for long term wearable monitoring. A novel variable input-feature correlated asynchronous sampling technique is proposed for direct sampling-cum-data compression at the ECG electrodes with embedded mixed-signal algorithm for direct recognition and capture of the fiducial points of the ECG signal, to indicate the occurrence of the Q, R and S waves. Simulation results show two key advantages (a) large compression at the source and (b) direct analog to information conversion with 98% accuracy of detection, with extremely low power and storage requirements.