A 1.06 μw smart ecg processor in 65 nm cmos for real-time biometric authentication and personal cardiac monitoring

A smart wearable electrocardiographic (ECG) processor is presented for secure ECG-based biometric authentication and cardiac monitoring, including arrhythmia and anomaly detection. Data-driven Lasso regression and low-precision techniques are developed to compress the neural networks by 24.4X. The prototype chip fabricated in 65 nm LP CMOS consumes 1.06 μW at 0.55 V for real-time ECG authentication. Equal error rates of 0.74% and 1.7% are achieved on ECG-ID database and in-house 645-subject database, respectively.