Pulse-Fi: A Low-Cost System for Accurate Heart Rate Monitoring Using Wi-Fi Channel State Information

Non-intrusive monitoring of vital signs has become increasingly important in various healthcare settings. In this paper, we present Pulse-Fi, a novel low-cost system that uses Wi-Fi Channel State Information (CSI) and machine learning to accurately monitor heart rate. Pulse-Fi operates using low-cost commodity devices, making it more accessible and cost-effective. It uses a signal processing pipeline to process CSI data fed into a custom low-compute Long Short-Term Memory (LSTM) neural network model. We evaluated Pulse-Fi using two datasets: one that we collected locally using ESP32 devices named ESP-HR-CSI Dataset and another containing recordings of 118 participants using the Raspberry Pi 4B called EHealth, making it the most comprehensive data set of its kind. Our results show that Pulse-Fi can effectively estimate heart rate from CSI signals with comparable or better accuracy than hardware with multiple antenna systems, which can be expensive.