Doppler Radar Heart Rate Sensor

ISEF Category: Embedded Systems

Subcategory: Sensors  ·  Difficulty: Intermediate  ·  Setup: Home Setup  ·  Time: 1 to 2 Months

The Hook

Your body already sends out motion signals strong enough for a radar to detect. That means a tiny sensor can track breathing and even heartbeat without touching your skin. The catch is movement. A good project here is all about separating tiny vital signs from bigger body shifts.

What Is It?

This project uses a 24 GHz Doppler radar module to detect motion from a person at a distance. Doppler radar sends out a signal and listens for the echo. When your chest moves from breathing or heartbeat, the echo shifts a little. Your ESP32 reads that change and turns it into a breathing or heart-rate estimate.

Think of it like listening to a whisper in a noisy room. Breathing creates a slow, large pattern. Heartbeat creates a much smaller, faster pattern. Body movement, blanket shifting, and posture changes add noise. Your job is to filter the useful signal from the rest and check how close your result comes to ECG ground truth.

Why This Is a Good Topic

This is a strong science fair topic because you can measure real performance, not just build a gadget. You can compare radar output against ECG data and ask how accuracy changes with distance, posture, motion, or filtering method. That gives you clear variables, clear controls, and a real engineering problem tied to sleep monitoring, home health, and contact-free sensing.

Research Questions

• How does subject distance from the radar affect heart-rate accuracy compared with ECG?
• What is the effect of body posture on breathing-rate detection quality?
• Does adaptive filtering improve heart-rate tracking during small voluntary movements?
• To what extent does blanket thickness change the radar signal for breathing detection?
• Which sensor placement gives the lowest error for non-contact respiration monitoring?
• How does the radar-based estimate compare with ECG across quiet and mildly active conditions?
• What is the effect of signal window length on the stability of the extracted heart rate?

Basic Materials

• 24 GHz hobby Doppler-radar module.
• ESP32 development board.
• Jumper wires and breadboard.
• USB cable for ESP32 programming.
• Laptop with Arduino IDE or PlatformIO.
• ECG chest strap or smartphone-compatible ECG device for ground truth.
• Stopwatch or timer app.
• Tripod, stand, or stable mount for the radar module.
• Tape measure for fixed subject distance.
• Notebook or spreadsheet for logging trials.

Advanced Materials

• 24 GHz radar module with access to raw I and Q outputs.
• ESP32 or similar microcontroller with enough memory for buffered sampling.
• ECG acquisition system with exportable time-series data.
• Oscilloscope or logic analyzer for signal debugging.
• Low-noise power supply or battery pack.
• Shielding materials or enclosure for interference tests.
• Reference motion platform or metronome-driven movement setup.
• MATLAB, Python, or similar analysis environment.
• Calibration target setup for repeatable distance testing.
• Environmental sensor for temperature, humidity, or motion context.

Software & Tools

• Arduino IDE: Programs the ESP32 and streams raw radar data for testing.
• Python: Cleans signals, runs filters, and computes accuracy metrics.
• Jupyter Notebook: Organizes trial-by-trial analysis and plots comparison graphs.
• Excel: Tracks protocol variables and helps check data quality quickly.
• ImageJ: Measures motion artifact patterns in plotted screenshots if you compare visual signal traces.

Experiment Steps

1. Define the exact vital sign you will measure first, then decide how you will label each trial with ground-truth ECG data.
2. Choose one motion artifact problem to attack first, such as posture shifts, small arm movement, or blanket interference.
3. Plan a clean baseline protocol so you can measure the radar in quiet conditions before adding harder real-world cases.
4. Design your signal-processing pipeline, including how you will separate breathing, heartbeat, and motion noise.
5. Build a comparison plan that turns raw radar output and ECG data into matching error metrics.
6. Set up fairness checks so every trial uses the same distance, alignment, and recording window.

Common Pitfalls

• Pointing the radar slightly off-axis, which weakens chest-motion return and makes the signal look unstable.
• Confusing heartbeat with breathing harmonics, which can make the extracted rate look correct by accident.
• Letting room motion or fan vibration contaminate the signal, which hides the true effect of your filter.
• Comparing radar and ECG recordings that are not time-synced, which inflates error for no real reason.
• Testing only one subject posture, which makes the results too narrow to support a strong conclusion.

What Makes This Competitive

A stronger version of this project goes past basic signal capture. You would compare multiple filtering methods, test more than one body position, and report error with clear statistics. You can also ask whether one model works better for breathing than for heart rate, since those signals live at different scales. That kind of careful analysis makes the project feel like real sensor research instead of a demo.

Project Variations

• Test the radar sensor through different bedding materials to see how fabric changes signal quality.
• Compare simple filtering, adaptive filtering, and machine learning based denoising on the same ECG-labeled dataset.
• Measure performance across sleep-like postures, such as supine, side-lying, and seated rest, to see which condition is easiest to track.

Learn More

• NIH PubMed: Search for review articles on contact-free vital sign monitoring, Doppler radar sensing, and motion artifact removal.
• NASA Open Courseware or MIT OpenCourseWare: Search for signal processing lectures that cover filtering, sampling, and frequency analysis.
• IEEE Xplore abstracts: Search for recent papers on radar-based respiration and heart-rate monitoring to compare methods and metrics.
• NOAA digital resources: Use background material on noise, waves, and signal interference concepts that help with sensor interpretation.
• Arduino Documentation: Read the ESP32 and signal input guides for basic hardware setup and data streaming examples.