Real-time heart rate variability detection on sensor node

Abstract

In this investigation we have prototyped a wireless ECG (electrocardiogram) sensor for real-time heart rate variability (HRV) detection, based on a commercially available sensor node. An ECG amplifier circuit is designed on a PCB board that is similar size to the sensor node so that it is easily attached to this node to achieve a minimum system size. In this implementation, the ECG is detected from both a patient simulator and a healthy subject body. The ECG signal is band-limited to 0.05Hz and 100Hz and amplified 1000 times before ADC conversion. We utilize the TinyOS 2.0.x open source operating system for providing data acquisition, computing and wireless communication services. A real-time HRV detection algorithm is implemented on the ECG sensor. The purpose is to reduce throughput of wireless data, and hence prolong battery life. Due to high sampling rate requirement (e.g 200Hz) of ECG detection, the HRV algorithms is required to complete within 5msec sample inter-arrival period in order to ensure no samples are lost. The digitized cardiac information is transmitted at 868 MHz ISM license free frequency band. A PC is used for displaying and storing the cardiac information.