The application of distributed radar to human motion monitoring is considered. A novel sensor fusion method has been proposed that yields a two-dimensional map of reflection intensity and a vector field of reconstructed velocities in lieu of conventional Doppler spectrograms or radial velocity components. The method has been verified using experimental datasets in two case studies involving fall detection in sequences of activities, and arm motion discrimination for in-place activities. A true positive rate and precision of respectively 99.3 % and 93.0 % have been demonstrated for the fall detection task, and the output of the proposed method for arm motion characterisation indicates suitability for classification in future research.

In this study, the problem of multipath in radar sensor networks for human activity recognition (HAR) has been examined. Traditionally considered as a source of additional clutter, the multipath is being investigated for its potential to be exploited through the creation of virtual radar nodes. These virtual nodes are conceptualized to observe targets from aspect angles that differ from those of physically existing radars. To realize this idea, an innovative processing pipeline is proposed that extracts information from multipath signals to improve HAR. The pipeline isolates and tracks the line-of-sight (LOS) and multipath components of a moving human target performing continuous sequences of activities observed by a network of three radar sensors. Furthermore, the method has been verified with experimental data consisting of six activities and 14 volunteers by comparing classification metrics with the use of a single radar as well as only the LOS components of the three radars in the network. A 12-layer convolutional neural network (CNN) classifier has been designed to operate on range-Doppler (RD) images derived from the LOS and multipath components, extracted by the proposed method. A substantial performance improvement using the leave-one-person-out (L1Po) test set is demonstrated in the order of +11% by exploiting a multiradar network with its LOS and multipath components.

Fall detection systems can play an important role in assuring safe independent living for vulnerable people. These sensors not only have to detect falls but also have to recognize uncritical, normal activities of daily living in order to differentiate them from falls. Radar sensors are very attractive for human activity recognition thanks to their contactless capabilities and lack of plain videos recorded. In this article, a novel approach to recognize single activities in a continuous stream of radar data is proposed, whereby the stream is divided into windows of fixed length and, then, multilabel classification is used to recognize all activities taking place in these time segments. While the initial feasibility of this approach was presented in an earlier contribution presented at the 2023 IEEE SENSORS conference, in this extended work, additional in-depth studies on critical parameters are performed. Specifically, multiple combinations of different radar data domains/representations (e.g., range-time maps, range-Doppler maps, and spectrograms) and different radar nodes in a network of five cooperating sensors are considered as inputs to two considered multilabel classification networks. In addition, a parametric study on the probability thresholds of the networks to assign labels to specific classes is also performed.

The effect of different time-frequency (TF) resolution values is analyzed in the context of Human Activity Recognition (HAR) using multiple radars distributed in a network. Specifically, different spectrograms computed with various Short-Time Fourier Transform (STFT) window lengths and Morse wavelet transform are compared as input representation to a Convolutional Neural Network (CNN), together with a coherent combination of multiple spectrograms. The study emphasizes the importance of selecting appropriate window sizes for TF analysis and for classification, balancing the observation time with the physical duration of the diverse activities, and also avoiding correlation between different data samples that may compromise the generalization ability of the method. The results employing this coherent sensor fusion demonstrate the efficacy of the investigated method, achieving an F1 score of 0.943 on a challenging public dataset containing 9 activities performed by 15 participants.

This paper presents a novel approach to radar-based human activity recognition in continuous data streams. To date, most work in this research area has aimed at either classifying every single time step separately by means of recurrent neural networks, or using a two-step procedure of first segmenting the stream into single activities and then classifying the segment. The first approach is restricted to time-dependent data as input; the second approach depends crucially on the segmentation step. To overcome these issues we propose a new approach in which we first segment the stream into windows of fixed length and subsequently classify each segment. Since due to the fixed length, the segment is not restricted to one activity alone, we use a multi-label classification approach, which can account for multiple activities taking place in the same segment by giving multiple outputs. To obtain a higher classification accuracy we fuse several radar data representations, namely range-time, range-Doppler and spectrogram. Using a publicly available dataset, an overall classification accuracy of 95.8% and F1 score of 92.08% could be achieved with the proposed method.

Classifying continuous sequences of human activities is a current research challenge due to the unconstrained duration of the constituent activities. Segmentation of these sequences into single-activity segments is under investigation as a potential solution to this challenge and has been studied in this work. A segmentation method has been proposed based on the extracted Rényi entropy of micro-Doppler spectrogram representations of human motion. The proposed method has been compared to a state of the art method for three different experimental data sets, for various sequence types, and in varying signal-to-noise regimes. It has been shown that the performance of the proposed method is up to 55 ± 22% higher than the reference method when applied to different data sets with unchanged parameters. Additionally, improved performance under degraded signal-to-noise ratio (SNR) conditions has been demonstrated for the proposed method. Finally, two methods for sensor fusion have been formulated for enhanced segmentation performance when multiple radar nodes are available, and have been demonstrated to increase performance by up to 10 ± 2%. The improved segmentation performance is expected to lead to improvements in continuous activity classification.