Smart wearable items are becoming more well recognized and are steadily making their way into people’s lives as a result of the ongoing advancement of technology and people’s growing concern for their health. In this work, we investigate the stimulation of physiological signals and the level of happiness indicated by people’s emotions using the linkage of smart gadgets and biological data. To reduce motion artifacts from wearable PPGs, we first suggest a sparse representation-based approach. To address the issue of poor model generalization brought on by individual signal differences (inter- and intra-individual) in human ECG data, a wearable ECG individual signal difference reduction technique based on Balanced Distribution Domain Adaptive (BDA) is also suggested. In addition to making a significant contribution to the advancement of intelligent control technology, medicine, and other fields, it provides an effective baseline for research on the satisfaction level of group music and dancing based on physiological signals.