The acquisition of PPG signal is vulnerable to the interference of motion artifact. In order to increase the accuracy of signal feature extraction and improve the accuracy of human physiological parameter calculation, a motion artifact filtering algorithm for PPG signal based on divergence value analysis is proposed. After the collected template signal and the experimental signal are preprocessed by band-pass filtering, the standard threshold range of the divergence value of the signal characteristics is calculated using the template signal that is not interfered by the motion artifacts. Then, the divergence value of the template signal and the experimental signal that is interfered by the motion artifacts is calculated, and the signal cycle with the motion artifacts interference in the experimental signal is identified, judged and eliminated, and the high-quality wave signal that is not interfered by the motion artifacts is integrated. The practicability and reliability of the algorithm in wearable sports system are proved through the detection experiment of human vascular systolic pressure.