An FFT-Based DC Offset Compensation and I/Q Imbalance Correction Algorithm for Bioradar Sensors
Fuze Tian, Lixian Zhu, Qiuxia Shi, Xiaokun Jin, Ran Cai, Qunxi Dong, Qinglin Zhao, Bin Hu
Abstract
The challenge of noncontact presentation of human cardiopulmonary activity using a bioradar sensor is to linearly demodulate the Doppler cardiopulmonary diagram (DCD) signal from baseband signals. Arctangent demodulation can perform linear phase demodulation to obtain the DCD signal. However, the high-order harmonics and intermodulation terms (ITs) caused by the time-varying direct current (dc) offset and in-phase and quadrature-phase (I/Q) imbalance in the baseband signals significantly degrade the signal-to-noise ratio (SNR) of the Doppler heartbeat diagram (DHD) signal. In this work, a fast Fourier transform (FFT)-based algorithm is proposed to simultaneously perform time-varying dc offset compensation and I/Q imbalance correction without the need for an auxiliary device to improve the accuracy of the arctangent demodulation. The obtained results show that the SNRs of the algorithm-processed DHD signals are increased from 30.08 <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\pm$</tex-math> </inline-formula> 2.41 to 68.88 <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\pm$</tex-math> </inline-formula> 10.57 dB. In addition, the root mean square errors (RMSEs) of the C-C intervals of the DHD signals for eight subjects with respect to the J-J intervals of the ballistocardiogram (BCG) signals are 17.79 <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\pm$</tex-math> </inline-formula> 2.72 ms (2.80% <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\pm$</tex-math> </inline-formula> 0.43%), suggesting a promising potential of the DHD signal for noncontact biomedical applications.