Gesture Control with 60 GHz Radar: Contactless Operation Without Machine Learning

12/03/2025 Radar Resources

In an era of heightened hygiene standards and intuitive interfaces, contactless operating concepts are becoming increasingly important. Radar provides a robust and precise solution that does not require cameras or machine learning.

A 60 GHz frequency-modulated continuous wave (FMCW) radar sensor can detect movements such as "swipe left", "swipe right", and "push" in order to control a motor. This technology is ideal for applications in medical technology, industry, and automotive systems because it recognizes gestures via radar.

The system architecture and signal processing

The Rutronik System Solutions team has developed a demonstrator to showcase the advantages of radar technology. At its core is a 60 GHz FMCW radar sensor from Infineon for motion detection. Signal processing is performed entirely on the Cortex-M55 core of the PSOC^TM Edge. Machine Learning was deliberately omitted to ensure the most robust and low-latency detection possible. This also eliminates the training effort typically required for ML applications.

The most important features of the demonstrator are listed below:

Sensor: 60 GHz radar IC BGT60TR13C from Infineon
MCU: Infineon PSOC^TM Edge with Cortex-M55
Interfaces: SPI, GPIO
Processing: Doppler FFT and azimuth calculation

Gesture recognition is ensured by analyzing magnitude, range, and azimuth. The graphic illustrates the signal processing from input to response.

The graphic illustrates the signal processing from input to response.

Input: Magnitude and angle of attack (AoA) as input data
Capturing: A 60 GHz FMCW radar sensor with one transmitting antenna and three receiving antennas provides a separate signal for each antenna
Processing: Fast Fourier transform (FFT) for motion detection
Detection: Phase differences of the receiving antennas determine direction
Radar control: Classification of the azimuth angle over time

Thanks to radar, the system is robust and reliable, and it is completely independent of lighting conditions, ambient noise, or varying hand positions.

With a latency of less than 10 milliseconds, the detection system responds almost in real time. It recognizes gestures such as "swipe left", "swipe right", and "push" precisely. The addition of the vertical dimension will allow it to integrate gestures such as "up" and "down" in the future.

Using 60 GHz radar for gesture recognition enables contactless operation, which is particularly useful in environments where hygiene is critical or where it is difficult to reach. Machine learning was deliberately omitted from 60 GHz radar gesture recognition, as it was not functionally necessary. This approach provides more robust, low-latency recognition without the need for training.

More information on radar gesture recognition with the BGT60TR13C and its corresponding evaluation kits can be found at rutronik.com/RAB3.

An article presenting a demonstrator for modern human-machine interaction using touch, speech, and gestures is available for free download here:

Discover more in the RUTRONIKER Article

Discover all Rutronik Radar Resources

Contactless operation with radar: Gesture recognition without Machine Learning