Measure heart rate with WiFi: Accurately monitor without wearing any devices.

Recently, an engineering team from the University of California, Santa Cruz, successfully measured heart rate using ordinary WiFi signals without the need for any wearable devices.

This technology, called "Pulse-Fi", uses low-cost WiFi chips combined with machine learning algorithms, achieving clinical-grade heart rate monitoring accuracy in just five seconds.

The researchers said that the system can accurately capture the heartbeat whether the person being measured is sitting, standing, lying down, or even moving around in the room.

What's most surprising is that Pulse-Fi remains stable in measurement even at a distance of three meters, or nearly ten feet.

In other words, an ordinary home WiFi router may be transformed into a health monitor in the future.

In the past, measuring heart rate always required close-fitting devices such as smartwatches, fitness rings, or hospital-grade equipment, but Pulse-Fi directly breaks this technological dependence.

The ESP32 chips used by the researchers cost only $5 to $10 each, far lower than traditional medical devices.

Even if the slightly more expensive Raspberry Pi chips are used, the total cost is only about $30, providing an attractive health monitoring solution for resource-scarce areas.

The research results have been published at the 2025 IEEE International Conference on Distributed Computing in Smart Systems and the Internet of Things.

01 Heartbeats in the Signals

The working principle of Pulse-Fi stems from a common phenomenon in every household: WiFi signals are absorbed and disturbed by surrounding objects.

WiFi devices continuously emit radio frequency waves into space. When these waves pass through obstacles such as the human body and furniture in the space, the waveform changes slightly.

The Pulse-Fi system consists of a transmitter and a receiver, and its core is a machine learning algorithm that can recognize heartbeat signals.

After training, the algorithm can isolate the tiny disturbances caused by the heartbeat from complex signals, eliminating the interference of environmental noise and human activities.

The researchers pointed out that the signals themselves are extremely sensitive. Without careful filtering, the traces of the heartbeat would be completely submerged in the background noise.

To train the algorithm, they had to collect signal samples themselves because no previous team had recorded such data using the ESP32 chip.

They built a test platform in the campus library, using a standard oximeter as the "ground truth" to synchronously record changes in the heartbeat and WiFi signals and construct a training dataset.

A total of 118 volunteers participated in the experiment, with each person tested in 17 different postures, resulting in more than 2,000 sets of data in total, providing rich samples for the model.

The team also referred to the world's largest WiFi heart rate dataset collected by a Brazilian research team using Raspberry Pi to further verify the universality of the algorithm.

The test found that when using Raspberry Pi devices, the system performance is even better than that of the ESP32, indicating that the higher the device grade, the more stable the measurement.

The key finding is that the measurement distance and posture have little impact on the performance. Even when the person being measured is far from the device and in various postures, the system still operates stably.

02 The Future Beyond Heartbeats

Currently, the researchers are promoting the next goal of Pulse-Fi: monitoring respiratory rate and even screening for respiratory abnormalities such as sleep apnea.

They have achieved "highly feasible" preliminary results in respiratory detection. Although the relevant papers have not been published yet, the experimental data is promising.

The project members said that one of the original design concepts of this technology is to cover all real-life scenarios, without requiring stillness or close proximity, as long as there is a WiFi signal.

This concept provides a solid foundation for building a "passive health monitoring system" in the future.

Imagine that the router in your living room could be more than just an internet access tool; it could be the "ear" of a family doctor.

This has great practical value for the elderly, patients with chronic diseases, and even sports enthusiasts.

In areas with scarce medical resources, this low-cost, non-contact technology will greatly relieve the burden on the medical system and expand the coverage of basic health services.

The core members of the research include computer engineering professor Katia Obraczka, doctoral student Nayan Bhatia, and visiting researcher Pranay Kocheta, who is only a high school student.

It's worth mentioning that this high school student not only participated in the modeling and algorithm verification but also promoted the design of multiple experiments and is an indispensable part of the project.

As the research progresses, this technology may change our daily lives and make "passive health monitoring" as common as WiFi.

Note: The cover image is AI-generated.

This article is from the WeChat official account "Big Data Digest", and is published by 36Kr with permission.