Search for a command to run...
In-car vital signs monitoring using millimeter-wave (mmWave) radar sensors has attracted growing interest due to its non-intrusive, real-time, and privacy-preserving nature. Leveraging the high-resolution capabilities of mmWave radar technology, this paper proposes non-contact detection of human presence and measurement of vital signs such as respiration and heartbeat within a vehicle cabin. A range of spectral analysis techniques—both parametric and nonparametric—are employed, including Fast Fourier Transform (FFT), Periodogram, Correlogram, ESPRIT (Estimation of signal parameters via rotational invariant techniques), MUSIC (Multiple Signal Classification), Nonlinear Least Squares (NLS), and the Iterative Adaptive Approach (IAA), to extract vital sign information with high precision. For real-time measurements, we utilize three radar platforms operating in the 60 GHz band: the BGT60LTR11AIP (Pulse-Doppler), the BGT60TR13C (SIMO), and the IWR6843ISK (MIMO). Experiments were conducted in diverse environments, including laboratory settings, on-road scenarios, and a custom test setup using a pump-embedded dummy to simulate infant vital signs. To distinguish living beings from inanimate objects, statistical features such as variance, entropy, and the Kolmogorov–Smirnov (KS) test are extracted and used as input to Support Vector Machine (SVM) and K-Nearest Neighbors (KNN) classifiers. The proposed solution is low-cost, privacy-preserving, and robust against environmental interference, making it ideal for integration into next-generation intelligent transportation systems.