Compressive sensing based ecg biometric system

Abstract

. The Internet of Things (IoT) has started redesigning the paradigm of the connected health sector by leveraging the availability of low power, low-cost sensors and efficient communication protocols. Consequently, IoT based connected health platforms are expected to further enhance the patient connectivity and everyday convenience. Nevertheless, issues related to power consumption and user security limit the performance of such systems. The conventional approaches that incorporate biometric measures into the IoT design rise high concerns regarding the cost and the complexity of the implementation. This paper proposes an identification approach integrated within a patient’s heart monitoring system based on the theory of compressive sensing (CS). CS is an emerging theory that promotes both power optimization and security by transmitting random measurements with fewer samples rather than transmitting the whole raw signal. The proposed system uses the electrocardiogram (ECG) as a biometric measure to identify the patient. The advantage of such system is that it does not require any additional complexity to acquire and process the data. The obtained results showed a successful identification rate up to 98.88% by compressing the transmitted signal to only half the original one.