The feasibility of classifying different human activities based on micro-Doppler signatures is investigated. Measured data of 12 human subjects performing seven different activities are collected using a Doppler radar. The seven activities include running, walking, walking while holding a stick, crawling, boxing while moving forward, boxing while standing in place, and sitting still. Six features are extracted from the Doppler spectrogram. A support vector machine (SVM) is then trained using the measurement features to classify the activities. A multiclass classification is implemented using a decision-tree structure. Optimal parameters for the SVM are found through a fourfold cross-validation. The resulting classification accuracy is found to be more than 90%. The potentials of classifying human activities over extended time duration, through wall, and at oblique angles with respect to the radar are also investigated and discussed.

We propose the use of deep convolutional neural networks (DCNNs) for human detection and activity classification based on Doppler radar. Previously, proposed schemes for these problems remained in the conventional supervised learning paradigm that relies on the design of handcrafted features. Whereas these schemes attained high accuracy, the requirement for domain knowledge of each problem limits the scalability of the proposed schemes. In this letter, we present an alternative deep learning approach. We apply the DCNN, one of the most successful deep learning algorithms, directly to a raw micro-Doppler spectrogram for both human detection and activity classification problem. The DCNN can jointly learn the necessary features and classification boundaries using the measured data without employing any explicit features on the micro-Doppler signals. We show that the DCNN can achieve accuracy results of 97.6% for human detection and 90.9% for human activity classification.

In this paper, we investigate the feasibility of recognizing human hand gestures using micro-Doppler signatures measured by Doppler radar with a deep convolutional neural network (DCNN). Hand gesture recognition using radar can be applied to control electronic appliances. Compared with an optical recognition system, radar can work regardless of light conditions and it can be embedded in a case. We classify ten different hand gestures, with only micro-Doppler signatures on spectrograms without range information. The ten gestures, which included swiping from left to right, swiping from right to left, rotating clockwise, rotating counterclockwise, pushing, double pushing, holding, and double holding, were measured using Doppler radar and their spectrograms investigated. A DCNN was employed to classify the spectrograms, with 90% of the data utilized for training and the remaining 10% for validation. After five-fold validation, the classification accuracy of the proposed method was found to be 85.6%. With seven gestures, the accuracy increased to 93.1%.

Adaptive matching methods for a wireless power transfer system in the near-field region are investigated. The impedance and resonant frequency characteristic of a near-field power transfer system are analyzed according to coupling distance. In the near-field region, adaptive matching is necessary to achieve an effective power transfer. We compare the power transfer efficiencies of several schemes including simultaneous conjugate matching and frequency tracking. It is found that effective adaptive matching can be easily achieved by tracking the split resonant frequency. In addition, a modified frequency tracking method is proposed to extend the range over which the power is transmitted with high efficiency. The experimental results are in agreement with the theoretical results.

We present a simulation methodology for generating micro-Doppler radar signatures of humans moving behind walls. The method combines primitive-based modeling of humans with finite-difference time-domain (FDTD) simulation of walls. Realistic motions of humans are generated from computer animation data. The time-varying human radar cross section is simulated using the primitive-based prediction technique. The scattered returns of humans behind walls are then simulated by a hybrid of the human simulation model with the through-wall propagation data generated from FDTD. The resulting simulator is used to investigate the effects of walls of both homogeneous and inhomogeneous types on human micro-Dopplers. It is found that while through-wall propagation affects the magnitude response of the Doppler spectrogram in the form of attenuation and fading, it only introduces very minor distortions on the actual Doppler frequencies from the body parts. This is corroborated by measurement data collected using a Doppler radar, as well as by a point-scatterer analysis of refraction and multipath introduced by walls.

In this letter, we propose a method for detecting a human subject using Doppler radar by investigating the physical characteristics of targets. Human detection has a number of applications in security, surveillance, and search-and-rescue operations. To classify a target from the Doppler signal, several features related to the physical characteristics of a target are extracted from a spectrogram. The features include the frequency of the limb motion, stride, bandwidth of the Doppler signal, and distribution of the signal strength in a spectrogram. The main contribution of this letter is the use of stride information of a target for the classification. Owing to the different lengths of legs and kinematic signatures of the target species, a human subject occupies a unique space in the domain of the stride and the frequency of limb motion. To verify the proposed method, we investigated humans, dogs, bicycles, and vehicles using the developed continuous-wave Doppler radar. The human subject is identified by a classifier of a support vector machine (SVM) trained to the extracted features. The trained SVM can detect a human subject with an accuracy of 96% with fourfold cross validation.