I/Q Imbalance Compensation and Block-Sparsity-Based Harmonic Analysis with Applications to Vital Signs Estimation using CW Radar Systems

Recently, breathing and heart rate estimation based on radar technologies have become increasingly popular. Millimeter wave radars are compact, low-cost and can perform vital signs estimation without the necessity of having physical contact with the subject under test. However, the strong harmonics of the breathing signal can contaminate the weak heart signal and make the estimation of the heart rate challenging. In this paper, we present the complete theory for vital signs estimation and address two challenging problems in this field, namely In-phase and Quadrature (I/Q) imbalance compensation as well as heart rate estimation in the presence of strong harmonics of the breathing signal. We address the heart rate estimation based on block sparsity model and cast the problem as a convex mixed-integer programming which can be solved efficiently. Finally, we p resent the experimental results gathered from a Continuous Wave (CW) radar operating at 24 GHz. The ground-truth signal to validate the overall accuracy and effectiveness of the algorithms is provided by the Electrocardiogram (ECG) device. Index Terms-I/Q imbalance compensation, block sparsity, breathing and heart rate estimation, convex mixed-integer programming, CW radars.