This paper proposes a system model and then verified by using the raw data for non-contact physiological sign monitoring using FMCW radar by integrating Empirical Mode Decomposition (EMD) and Continuous Wavelet Transform (CWT). The model effectively extracts respiratory and cardiac rates from modeled radar signals and the result is validated by the raw data obtained from FMCW radar IWR 1843 BOOST by Texas Instrument. The approach addresses challenges such as noise, baseline drift, and low-frequency resolution through adaptive filtering, enhanced IMF selection with frequency stability analysis, and high-resolution spectral estimation. The modeling results demonstrate an MAE of 3.8 and 90.7 mHz of respiration and heart rates across a range of physiological frequencies, validating the potential of the EMDCWT framework for real-time radar-based health monitoring applications.