A singular spectrum analysis based human life signal detection

Life signal detection of human beings behind an obstacle is a keen research topic in recent years. In this paper, a singular spectrum analysis (SSA) based human life signal detection approach is proposed. SSA is a nonparametric spectral estimation method, and is widely used in time series analysis. SSA can decompose a time series into a sum of components, such as trend, periodic components and noise, so it is more suitable to detect human life signal. Firstly, stationary human micro-motion model is established, and the received signal of continuous wave radar is analyzed. Then SSA is used to detect human life signal by four steps: embedding, singular value decomposition, grouping and reconstruction. Simulation and experimental results validate the correctness and effectiveness of SSA in human life signal detection.