Personal sleep control monitoring nowadays is gaining popularity both thanks to rapid development of compact and cheap sensors and to increase of understanding of human movements behavior during sleep. One relevant direction is monitoring of periodic limb movement disorder (PLMD). PLMD is a sleep disorder, which is characterized by repetitive jerking, cramping, or twitching of lower limbs during sleep, and can be symptom of diabetes, iron deficiency, uremia, anemia, etc. With appearance of compact and accurate ultra wideband radars, PLMD monitoring became available for users at home environment. In this paper, we propose an algorithm of PLMD monitoring during sleep using off-the-shelf UWB radar. The applicability and accuracy of the method was verified at clinical tests with 110 subjects. Proposed algorithm achieves 0.86 F1-score.

In this work, we demonstrated the results of experimental trials of a bio-electrical impedance analysis (BIA) for smartwatches that is robust to high skin-contact impedances and poorly predictable parasitic impedance values. This makes the BIA process independent of the complex hardware of smartwatches, including the small size of electrodes, and avoids the need for additional factory-mode calibrations because of hardware bias. The applicability and accuracy of the method were verified through clinical trials with 90 participants using pre-commercial prototype smartwatch with two ~0.3 cm2area and two ~1.5 cm2area electrodes. The proposed algorithm has achieved a similar performance in terms of the body fat mass estimation (R2is 94.3% and RMSE is 2.22 kg) compared to a professional medical device (R2is 94.6% and RMSE is 2.17 kg). Body fat mass measured by smartwatches can help users monitor their fat percentage over a longer period and make informed of deviations from normal levels.

We propose the extension of the smartwatches functionality by high performance algorithm of the abdomen visceral fat area (VFA) estimation. VFA is a significant risk factor for the development of diseases and have a significant impact on the life quality. Algorithm has been verified at clinical trials with 90 subjects at uniform distribution of the age and body mass indexes. Proposed algorithm reaches the high performance in terms of VFA estimation metrics (coefficient of determination (R2) is 0.90 and root mean square error is 14.6cm2). Solution provides new experience for user to monitor his VFA in convenient way via ecosystem of the personal connected devices, like smartwatch and smartphone.

This paper presents a novel approach for the development of compact 360° scanning antennas. The proposed antenna represents a cylindrical end-fire antenna array designed for 57-71GHz frequency range. In combination with an advanced approach of non-galvanic interconnection, this antenna provides a hemispherical coverage using a minimum number of active RF channels with no additional switching components. This is strongly required, for example, for AR Glasses and can be used in other devices, including embedded in the vehicle, operated in 5G, Beyond 5G, WiGig and 6G. Concept is successfully verified in the experiments.

This paper presents a design of mm Wave antenna array unit for imaging radar to determine the position of an object in the three-dimensional space. Proposed design gives the following advantages in comparison with existing solutions: high field-of-view to resolution ratio for the radar at minimum transceiver numbers, smallest possible sensor size in both elevation and azimuth plane, multi-functional support for both objects tracking and objects imaging. Antenna array unit provides supports 4 transmitting and 4 receiving channels (forming 16 virtual channels). Field of view is 21° and resolution is 10.5° in elevation plane. Field of view is 80° and resolution is 10.8° in azimuth plane. (Abstract)

This paper presents an ultra-wide scanning topology of the multiple-input-multiple-output antenna array for Imaging Radar to be the basis for algorithms development in future implementation of AI-driven cars. Antenna array provides the ultra-wide field of view (145°) at 2.3° resolution in azimuth plane and was developed for a 79 GHz frequency modulated continuous wave radar prototype for use in autonomous vehicles (Level-4) and supports 12 transmitting and 16 receiving channels (forming 192 virtual channels). Field of view is 20° and resolution is 2.75° in elevation plane. Antenna array was installed on the vehicle as part of the Imaging Radar system. The basic performances were evaluated at field tests.

This paper presents a method of increasing isolation up to 45dB between close located$(\sim\lambda/2)$antenna array elements. Proposed method allows to minimize the spillover between adjacent antenna elements. It is important for compact topology design of the multiple-input-multiple-output antenna array for high-resolution automotive radar required to support of a large number of transmitters and receivers. Proposed method was realized on the base of two-resonance stripline-fed aperture-coupled antenna element and deliver$> 40\text{dB}$isolation level for 76–81 GHz frequency band. Antenna array was developed for a 79 GHz frequency modulated continuous wave radar prototype for use in autonomous vehicles (Level-4) and supports 12 transmitting and 16 receiving channels (forming 192 virtual channels). Antenna array was installed on the vehicle as part of the Imaging Radar system. The basic performances were evaluated at field tests.

A 79-GHz frequency modulated continuous wave (FMCW) radar prototype for use in autonomous vehicles (Level-4), and its basic performance was evaluated. The system comprised 12 Tx and 16 Rx multiple-input-multiple-output (MIMO) antennas with high resolution in both the horizontal and vertical directions. The system was loaded on the vehicle, and the main functions, such as detection, tracking, and classification, were assessed via a road driving test. Furthermore, a grid map radar image was implemented to image a parked vehicle next to the road.