This work focuses on a radar-based fall detection technique. In order to address the fall detection problem, a new method to analyze the radar returns is proposed based on time-frequency analysis and deep neural network. The time-frequency analysis is first performed by applying the short-time Fourier transform and resulting spectrogram is projected onto a binary image representation. Unlike most of the existing works, the proposed method is realized by a high-level feature learning from radar return signals using deep neural network. This method adopts a deep convolutional neural network to automate feature extraction and accurately classifies fall and non-fall activities. The results show that the proposed method is capable of distinguishing fall from non-fall activities with low false alarm rate. In the proposed method, hierarchical features extracted from radar returns are the key points to understanding fundamental composition of human activities and determining whether or not a fall has occurred.

Automatic fall detection using radar aids in better assisted living and smarter health care. In this brief, a novel time series-based method for detecting fall incidents in human daily activities is proposed. A time series in the slow-time is obtained by summing all the range bins corresponding to fast-time of the ultra wideband radar return signals. This time series is used as input to the proposed deep convolutional neural network for automatic feature extraction. In contrast to other existing methods, the proposed fall detection method relies on multi-level feature learning directly from the radar time series signals. In particular, the proposed method utilizes a deep convolutional neural network for automating feature extraction as well as global maximum pooling technique for enhancing model discriminability. The performance of the proposed method is compared with that of the state-of-the-art such as recurrent neural network, multi-layer perceptron and dynamic time warping techniques. The results demonstrate that the proposed fall detection method outperforms the other methods in terms of higher accuracy, precision, sensitivity and specificity values.

In this study, a contactless vital signs monitoring system was proposed, which can measure body temperature (BT), heart rate (HR) and respiration rate (RR) for people with and without face masks using a thermal and an RGB camera. The convolution neural network (CNN) based face detector was applied and three regions of interest (ROIs) were located based on facial landmarks for vital sign estimation. Ten healthy subjects from a variety of ethnic backgrounds with skin colors from pale white to darker brown participated in several different experiments. The absolute error (AE) between the estimated HR using the proposed method and the reference HR from all experiments is $2.70\pm 2.28$ beats/min (mean ± std), and the AE between the estimated RR and the reference RR from all experiments is $1.47\pm 1.33$ breaths/min (mean ± std) at a distance of 0.6–1.2 m.

Radar-based noncontact sensing of life sign signals is often used in safety and rescue missions during disasters such as earthquakes and avalanches and for home care applications. The radar returns obtained from a human target contain the breathing frequency along with its strong higher harmonics depending on the target’s posture. As a consequence, well understood, computationally efficient, and the most popular traditional FFT-based estimators that rely only on the strongest peak for estimates of breathing rates may be inaccurate. The paper proposes a solution for correcting the estimation errors of such single peak-based algorithms. The proposed method is based on using harmonically related comb filters over a set of all possible breathing frequencies. The method is tested on three subjects for different postures, for different distances between the radar and the subject, and for two different radar platforms: PN-UWB and phase modulated-CW (PM-CW) radars. Simplified algorithms more suitable for real-time implementation have also been proposed and compared using accuracy and computational complexity. The proposed breathing rate estimation algorithms provide a reduction of about 81% and 80% in the mean absolute error of breathing rates in comparison to the traditional FFT-based methods using strongest peak detection, for PN-UWB and PM-CW radars, respectively.

As a technology that can accommodate more users and significantly improve spectral efficiency, non-orthogonal multiple access (NOMA) has attracted the attention of many scholars in recent years. The basic idea of NOMA is to implement multiple access in the power domain and decode the desired signal via successive interference cancellation (SIC). However, the resource allocation problem in such NOMA system is non-convex. It is difficult to directly solve this optimization problem through conventional methods. As such, we propose to apply a reinforcement learning (RL) approach based on cooperative Q-learning to solve the resource allocation problem in multi-antenna downlink NOMA systems. First, we formulate the resource allocation process as a sum rate maximization problem, subject to the power budget constraints and quality of service (QoS) condition. Second, we design a reward function to improve the sum rate while meeting the power and capacity constraints. Multiple Q-tables are created and cooperatively updated to get the optimal beamforming matrix. Then, we analyze the convergence of our proposed RL based power allocation method. Our simulations show that the proposed power allocation scheme yields excellent performance in terms of sum rate, energy efficiency, and spectral efficiency.

Recombinant adeno-associated virus (rAAV) is the most effective viral vector technology for directly translating the genomic revolution into medicinal therapies. However, the manufacturing of rAAV viral vectors remains challenging in the upstream processing with low rAAV yield in large-scale production and high cost, limiting the generalization of rAAV-based treatments. This situation can be improved by real-time monitoring of critical process parameters (CPP) that affect critical quality attributes (CQA). To achieve this aim, soft sensing combined with predictive modeling is an important strategy that can be used for optimizing the upstream process of rAAV production by monitoring critical process variables in real time. However, the development of soft sensors for rAAV production as a fast and low-cost monitoring approach is not an easy task. This review article describes four challenges and critically discusses the possible solutions that can enable the application of soft sensors for rAAV production monitoring. The challenges from a data scientist’s perspective are (i) a predictor variable (soft-sensor inputs) set without AAV viral titer, (ii) multi-step forecasting, (iii) multiple process phases, and (iv) soft-sensor development composed of the mechanistic model.