Test Signal Generation and Algorithm Testing for Wearable Photoplethysmography

Wearable photoplethysmography (PPG) signals are often degraded by motion and environmental interference, making algorithm validation difficult when relying solely on experimental datasets. This paper presents a PPG waveform synthesizer that generates realistic traces with controlled physiological variability, baseline drift, noise, and two classes of motion artifacts. The model is based on a two-Gaussian pulse description with stochastic heart-rate variability and provides full ground truth for pulse timing and artifact locations. We use the synthesizer to evaluate two low-complexity heartrate estimation methods - a peak-tracking algorithm derived from the Mountaineer's method and a Sliding Discrete Fourier Transform (SDFT) approach - under diverse simulated conditions. Results show sub-bpm accuracy in clean and mildly corrupted traces, while uncorrelated waveform artifacts distinctly reveal algorithm robustness limits. The proposed synthesizer enables reproducible, large-scale testing and supports the design of reliable algorithms for wearable PPG monitoring in everyday activity.