A simple pulse analyzer for device with limited computational power

In this paper, we present a simple approach for implementing a pulse analyzer for devices with limited computational power, such as mobile and wearable computers. The main purpose of this system is to continuously monitor the ECG of a person and detect any possible heart abnormalities. The idea is to break the required computation in two parts. The learning and model building part that requires high computation is done on a desktop computer. Once the model for a person is built, it uses a low computational cost signal analysis approach to analyze the signals in real-time. The system has been implemented on an IBM WatchPad, a wearable Linux computer designed as a watch. The experiments conducted show that the system produces reliable results. This approach can benefit the embedded systems in the medical domain, such as the personal health monitoring devices, etc.