A Running Heart Rate Prediction Model Based on Variational Mode Decomposition and TCN

Predicting heart rate during running is crucial for personalized exercise programs and health management. However, existing studies are deficient in handling non-smooth signals and long time dependency, which affects the accuracy and stability of prediction. In this study, we propose an innovative heart rate prediction model based on variational modal decomposition and temporal convolutional network. Variational Mode Decomposition decomposes complex heart rate signals into multiple intrinsic modal functions to efficiently capture multi-scale features, which solves the problem of insufficient processing of non-smoothed signals. Temporal Convolutional Network utilizes causal convolution and extended convolution techniques to efficiently capture long time dependencies. In the experimental results results show that the accuracy of the proposed model in heart rate prediction is better than other models.