Accurate Person Identification Based on Combined Sit-to-Stand and Stand-to-Sit Movements Measured Using Doppler Radars

Abstract

This paper demonstrates the identification of 10 persons with 99% accuracy achieved by combining micro-Doppler signatures of sit-to-stand and stand-to-sit movements. Data from these movements are measured using two radars installed above and behind the person. Images of Doppler spectrograms generated using the measured data are combined and input to a convolutional neural network. Experimental results show the significantly better accuracy of the proposed method compared with conventional methods that do not perform data combination. The accuracy of identifying 10 participants having similar ages and physical features was 96-99%, despite the relatively small training set (number of training samples: 50-90 Doppler radar images per person). These results suggest that combining sit-to-stand and stand-to-sit movements provides sufficient information for accurate person identification and such information can be remotely acquired using Doppler radar measurements.

A systems approach towards remote health-monitoring in older adults: Introducing a zero-interaction digital exhaust

“…Finally, one drawback of simple PIR and door sensors is that not all digital measures based on this technologies can be calculated when more than one person is living in an apartment. Here a potential solution may be found in wireless radio signal technologies 14 , 72 , 73 , that could not only provide more accurate digital measures, compared to simple PIR motion sensors, but should also be able to differentiate between multiple persons, by detecting specific signatures (such as gait characteristics) 74 .…”

Personal Identification Based on the Picking‐Up Movement Measured Using a Doppler Radar

“…The Doppler radar can achieve an accurate walking (gait)-based identification using deep learning of the time-velocity distribution (spectrogram) with an accuracy of above 90% [4,5]. Furthermore, in our previous study, we achieved an accurate person identification with 99% accuracy based on the standingup and sitting-down movements [6]. Lang et al report a person identification capability using other motions such as running, jumping, boxing, and creeping with over 75% accuracy, even with limited training data [5].…”

“…The spectrograms were calculated using a short-time Fourier transform using a Hamming window with a length of 128 samples and an overlap of 127 samples. similar studies [4][5][6]. The generated spectrograms were converted to PNG images of 224 × 224 pixel-size.…”

“…The length for all the data measurements was 3.3 s, which is sufficient to record the pickingup movement. The architecture of the CNN was ResNet-18, which is an effective network for a radar-based person or motion recognition used in previous studies [5,6]. The neural network was trained using 70% of the generated images (35 spectrograms for each participant).…”

“…However, its application is difficult in cases where a person wears a mask, and the reference face image is not sufficiently clear. To solve this problem, a promising approach is personal identification based on the body movements and behaviors, such as walking, standing-up, and sitting-down [2][3][4][5][6].…”