Radar sensors offer a promising and effective sensing modality for human activity classification. Human activity classification enables several smart homes applications for energy saving, human-machine interface for gesture controlled appliances and elderly fall-motion recognition. Present radar-based activity recognition system exploit micro-Doppler signature by generating Doppler spectrograms or video of range-Doppler images (RDIs), followed by deep neural network or machine learning for classification. Although, deep convolutional neural networks (DCNN) have been shown to implicitly learn features from raw sensor data in other fields, such as camera and speech, yet for the case of radar DCNN preprocessing followed by feature image generation, such as video of RDI or Doppler spectrogram, is required to develop a scalable and robust classification or regression application. In this paper, we propose a parametric convolutional neural network that mimics the radar preprocessing across fast-time and slow-time radar data through 2D sinc filter or 2D wavelet filter kernels to extract features for classification of various human activities. It is demonstrated that our proposed solution shows improved results compared to equivalent state-of-art DCNN solutions that rely on Doppler spectrogram or video of RDIs as feature images.

This paper presents a parametric variational autoencoder-based human target detection and localization framework working directly with the raw analog-to-digital converter data from the frequency modulated continuous wave radar. We propose a parametrically constrained variational autoencoder, with residual and skip connections, capable of generating the clustered and localized target detections on the range-angle image. Furthermore, to circumvent the problem of training the proposed neural network on all possible scenarios using real radar data, we propose domain adaptation strategies whereby we first train the neural network using ray tracing based model data and then adapt the network to work on real sensor data. This strategy ensures better generalization and scalability of the proposed neural network even though it is trained with limited radar data. We demonstrate the superior detection and localization performance of our proposed solution compared to the conventional signal processing pipeline and earlier state-of-art deep U-Net architecture with range-doppler images as inputs.

No abstract

Radar sensors offer a promising and effective sensing modality for<br>human activity classification. Human activity classification enables several smart<br>homes applications for energy saving, human-machine interface for gesture<br>controlled appliances and elderly fall-motion recognition. Present radar-based<br>activity recognition system exploit micro-Doppler signature by generating Doppler<br>spectrograms or video of range-Doppler images (RDIs), followed by deep neural<br>network or machine learning for classification. Although, deep convolutional neural<br>networks (DCNN) have been shown to implicitly learn features from raw sensor<br>data in other fields, such as camera and speech, yet for the case of radar DCNN<br>preprocessing followed by feature image generation, such as video of RDI or<br>Doppler spectrogram, is required to develop a scalable and robust classification<br>or regression application. In this paper, we propose a parametric convolutional<br>neural network that mimics the radar preprocessing across fast-time and slow-time<br>radar data through 2D sinc filter or 2D wavelet filter kernels to extract features for<br>classification of various human activities. It is demonstrated that our proposed<br>solution shows improved results compared to equivalent state-of-art DCNN solutions<br>that rely on Doppler spectrogram or video of RDIs as feature images.

The paper proposes a data-driven pre-processing optimization for radar data using a parametric convolutional neural network. The proposed method is applied on human activity classification as a use case. Present radar-based activity recognition system exploit micro-Doppler signature by generating Doppler spectrograms or a temporal series of range-Doppler maps, followed by deep neural networks or machine learning approaches for classification. Those radar data representations are typically generated on the basis of short-time Fourier transformations. A Fourier transformation equally resolves the frequency space, which may be sub-optimal in some applications. Although deep convolutional neural networks (DCNN) have been shown to implicitly learn features from raw sensor data in other fields, such as speech recognition, yet, for the case of radar-based DCNNs, pre-processing is required to develop a scalable and robust classification or regression application. In this paper, we propose a parametric convolutional neural network that mimics the radar pre-processing across fast-time and slow-time radar data through 2D sinc filter or 2D wavelet filter kernels to extract features for classification of various human activities. During training only the filter parameters of the 2D sinc filters or 2D wavelets are learned, leading to optimized feature representation for the classification task. It is demonstrated that our proposed solution shows improved results compared to equivalent DCNN architectures that rely on Doppler spectrograms or radar data cubes as input data.