Contactless Vital Signal Detection Method Based on TVF-EMD and VMD

With the development of contactless vital sign detection technology, it has demonstrated broad application prospects in fields such as medical monitoring and post-disaster search and rescue. However, the amplitude of heartbeat signals is extremely weak, even less than one-tenth of MZ signals. In high-noise environments, traditional algorithms are prone to interference from signal harmonics and environmental noise, making it difficult to effectively separate heartbeat signals with a high signal-to-noise ratio (SNR). To address this issue, this paper proposes a signal processing method that integrates Time-Varying Filtering Empirical Mode Decomposition (TVF-EMD) and Variational Mode Decomposition (VMD). First, TVF-EMD is used to extract vital sign intrinsic mode functions (IMFs) from noisy vital sign signals. Then, VMD is applied to achieve an accurate extraction of the heartbeat signal. This paper simulates radar-received vital sign signals, extracts heartbeat signals using TVF-EMD and VMD, and conducts comparative experiments with EMD, VMD, TVF-EMD, as well as fusion algorithms such as TVF-EMD+EMD and SSA+VMD. The experimental results show that the proposed method outperforms other algorithms in both SNR and correlation coefficient, with the maximum SNR improvement exceeding $50 \%$. These results validate the effectiveness of the proposed method.