How do smartwatch heart rate sensors work?

27 Tem 23:16

Optical PPG (photoplethysmography) sensors in most smartwatches work by measuring changes in the volume of blood vessels under the skin. An LED emits light at a specific wavelength (usually green, sometimes infrared), a photodetector captures the reflected light, and an algorithm converts the amplitude fluctuations into heart rate. Some models, like the Apple Watch Series 4 and newer, supplement PPG with an electrocardiographic (ECG) module, which measures the heart's electrical potential through electrode contacts on the case and band. Data processing happens in real time on a microcontroller: first, noise filtering is applied (bandpass and blink filters), then RR intervals are calculated, and BPM is derived from them. To reduce motion interference, adaptive algorithms "select" reliable signal segments and ignore artifacts, while also using an accelerometer for correction. However, issues do arise during intense movement, dark or very light skin tones, and in conditions of bright external lighting. How effective do you think current compensation methods (such as multispectral light or factory calibration) are at handling these scenarios in real-world use? Perhaps adding extra sensors (like temperature sensors) could help stabilize the signal. If you’ve had hands-on experience with a specific model and noticed the algorithm "failing" during certain activities (e.g., running with a high step frequency), please share the details—it’ll help identify where improvements are most needed.