Search for: All records

Identity authentication based on Doppler radar respiration sensing is gaining attention as it requires neither contact nor line of sight and does not give rise to privacy concerns associated with video imaging. Prior research demonstrating the recognition of individuals has been limited to isolated single subject scenarios. When two equidistant subjects are present, identification is more challenging due to the interference of respiration motion patterns in the reflected radar signal. In this research, respiratory signature separation techniques are functionally combined with machine learning (ML) classifiers for reliable subject identity authentication. An improved version of the dynamic segmentation algorithm (peak search and triangulation) was proposed, which can extract distinguishable airflow profile-related features (exhale area, inhale area, inhale/exhale speed, and breathing depth) for medium-scale experiments of 20 different participants to examine the feasibility of extraction of an individual’s respiratory features from a combined mixture of motions for subjects. Independent component analysis with the joint approximation of diagonalization of eigenmatrices (ICA-JADE) algorithm was employed to isolate individual respiratory signatures from combined mixtures of breathing patterns. The extracted hyperfeature sets were then evaluated by integrating two different popular ML classifiers, k-nearest neighbor (KNN) and support vector machine (SVM), for subject authentication. Accuracies of 97.5% for two-subject experiments and 98.33% for single-subject experiments were achieved, which supersedes the performance of prior reported methods. The proposed identity authentication approach has several potential applications, including security/surveillance, the Internet-of- Things (IoT) applications, virtual reality, and health monitoring. 

A number of algorithms have been developed to extract heart rate from physiological motion data using Doppler radar system. Yet, it is very challenging to eliminate noise associated with surroundings, especially with a single-channel Doppler radar system. However, single-channel Doppler radars provide the advantage of operating at lower power. Additionally, heart rate extraction using single-channel Doppler radar has remained somewhat unexplored. This has motivated the development of effective signal processing algorithms for signals received from single-channel Doppler radars. Three algorithms have been studied for estimating heart rate. The first algorithm is based on applying FFT on an FIR filtered signal. In the second algorithm, autocorrelation was performed on the filtered data. Thirdly, a peak finding algorithm was used in conjunction with a moving average preceded by a clipper to determine the heart rate. The results obtained were compared with heart rate readings from a pulse oximeter. With a mean difference of 2.6 bpm, the heart rate from Doppler radar matched that from the pulse oximeter most frequently when the peak finding algorithm was used. The results obtained using autocorrelation and peak finding algorithm (with standard deviations of 2.6 bpm and 4.0 bpm) suggest that a single channel Doppler radar system can be a viable alternative to contact heart rate monitors in patients for whom contact measurements are not feasible. 

Measurement of the body's displacement at multiple positions allows heart pulse wave propagation to be observed; this is an important step toward noncontact blood pressure measurement. This study investigates the feasibility of performing blood pressure measurements using skin displacement waveforms measured at two positions on a human body. To evaluate the accuracy of the proposed approach, this study uses a pair of laser displacement sensors to enable precise pulse transit time measurement. By comparing the displacement waveforms from the two sensors, the relationship between pulse transit time and blood pressure was evaluated. It is demonstrated experimentally that the blood pressure can be estimated with accuracy of 5.1 mmHg, which is equivalent to the error of an ordinary cuff-type blood pressure monitor.