We propose a drone classification method based on convolutional neural network (CNN) and micro-Doppler signature (MDS). The MDS only presents Doppler information in time domain. The frequency domain representation of MDS is called as cadence-velocity diagram (CVD). To analyze the Doppler information of drone in time and frequency domain, we propose a new image by merging MDS and CVD, as merged Doppler image. GoogLeNet, a CNN structure, is utilized for the proposed image data set because of its high performance and optimized computing resources. The image data set is generated by the returned Ku-band frequency modulation continuous wave radar signal. Proposed approach is tested and verified in two different environments, anechoic chamber and outdoor. First, we tested our approach with different numbers of operating motor and aspect angle of a drone. The proposed method improved the accuracy from 89.3% to 94.7%. Second, two types of drone at the 50 and 100 m height are classified and showed 100% accuracy due to distinct difference in the result images.

This paper proposes a method to simultaneously detect and classify objects by using a deep learning model, specifically you only look once (YOLO), with pre-processed automotive radar signals. In conventional methods, the detection and classification in automotive radar systems are conducted in two successive stages; however, in the proposed method, the two stages are combined into one. To verify the effectiveness of the proposed method, we applied it to the actual radar data measured using our automotive radar sensor. According to the results, our proposed method can simultaneously detect targets and classify them with over 90% accuracy. In addition, it shows better performance in terms of detection and classification, compared with conventional methods such as density-based spatial clustering of applications with noise or the support vector machine. Moreover, the proposed method especially exhibits better performance when detecting and classifying a vehicle with a long body.

We present a polarimetric analysis of small drones from different aspect angles. Polarimetric analysis can provide more information of a target, since the returned radar signal is affected by different wave polarization. The analysis is performed with micro-Doppler signature (MDS) to investigate micromotions of the drone detected by the radar. We measured operating small drones in an anechoic chamber from two aspect angles, 0° and 90°. An outdoor experiment was carried out with metal clutters for verification in real environment. The indoor analysis result shows that copolarized antenna receives signals better than cross polarized when the aspect angle is 0°, and vice versa. We also verified that cross-polarized antenna receives MDS from the drone better than copolarized antenna, from outdoors when an aspect angle is almost 90°. By utilizing the polarimetric characteristic of the drone at this frequency band, it is preferable to use a polarimetric radar for drone detection.

We propose a drone classification method for polarimetric radar, based on convolutional neural network (CNN) and image processing methods. The proposed method improves drone classification accuracy when the micro-Doppler signature is very weak by the aspect angle. To utilize received polarimetric signal, we propose a novel image structure for three-channel image classification CNN. To reduce the size of data from four different polarization while securing high classification accuracy, an image processing method and structure are introduced. The data set is prepared for a three type of drone, with a polarimetric Ku-band frequency modulated continuous wave (FMCW) radar system. Proposed method is tested and verified in an anechoic chamber environment for fast evaluation. A famous CNN structure, GoogLeNet, is used to evaluate the effect of the proposed radar preprocessing. The result showed that the proposed method improved the accuracy from 89.9% to 99.8%, compared with single polarized micro-Doppler image. We compared the result from the proposed method with conventional polarimetric radar image structure and achieved similar accuracy while having half of full polarimetric data.

In this letter, we propose a method of monitoring driver's movement using a small-sized millimeter-wave radar sensor. First, we install a frequency-modulated continuous wave radar sensor on the steering wheel and collect reflected signals for various driver motions. Because the reflection patterns are different for each motion, we can distinguish different movements of the driver by analyzing the received signal. In our method, we use the spectrogram of the received signal obtained through multiple measurements and apply a window function to extract the radar signal in the form of an image. Then, we use the convolutional neural network architecture to classify the radar signal images. The classification results show that our proposed method can classify four different movements of the driver with an accuracy higher than 80%.

This paper presents a novel method to reduce measurement time for remote heart rate measurement using mmWave FMCW radar. The proposed method is based on the sampling theory and the frame structure of recent FMCW radar systems. The range displacement from heartbeat is measured by 60GHz FMCW radar with a fixed frame interval, 50ms. With additional chirp repetition in each frame, the heartbeat signal can be sampled with increased or decreased interval. By adopting three different sampling rates from a single radar measurement, the resolution of heart rate measurement is improved in a short time, 2.5s. The proposed method is evaluated with a commercial off-the-shelf FMCW radar system and three different human subject, for no respiration and natural status. To evaluate the performance of the proposed method, the heart rate result is compared with ECG sensor and conventional signal processing method. The results show that the proposed method estimates the heart rate with lower error and shorter measurement time.