Renyuan Zhang

Renyuan Zhang

orcid

Research Focus & Metrics

Academic Career Length

10Years (2017-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 September 2020 in IEEE Sensors Journal
Verified
Journal Article
Closed access
ASArindam Sengupta
Arindam Sengupta
FJFeng Jin
Feng Jin
RZRenyuan Zhang
Renyuan Zhang
SCSiyang Cao
Siyang Cao
In this paper, mm-Pose, a novel approach to detect and track human skeletons in real-time using an mmWave radar, is proposed. To the best of the authors’ knowledge, this is the first method to detect >15 distinct skeletal joints using mmWave radar reflection signals. The proposed method would find several applications in traffic monitoring systems, autonomous vehicles, patient monitoring systems and defense forces to detect and track human skeleton for effective and preventive decision making in real-time. The use of radar makes the system operationally robust to scene lighting and adverse weather conditions. The reflected radar point cloud in range, azimuth and elevation are first resolved and projected in Range-Azimuth and Range-Elevation planes. A novel low-size high-resolution radar-to-image representation is also presented, that overcomes the sparsity in traditional point cloud data and offers significant reduction in the subsequent machine learning architecture. The RGB channels were assigned with the normalized values of range, elevation/azimuth and the power level of the reflection signals for each of the points. A forked CNN architecture was used to predict the real-world position of the skeletal joints in 3-D space, using the radar-to-image representation. The proposed method was tested for a single human scenario for four primary motions, (i) Walking, (ii) Swinging left arm, (iii) Swinging right arm, and (iv) Swinging both arms to validate accurate predictions for motion in range, azimuth and elevation. The detailed methodology, implementation, challenges, and validation results are presented.

21 December 2018 in IEEE Sensors Letters
Verified
Journal Article
Closed access
RZRenyuan Zhang
Renyuan Zhang
SCSiyang Cao
Siyang Cao
A real-time motion behavior detection system using millimeter wave (mmWave) radar is presented in this paper. Radar is used to sense the micro-Doppler information of targets. A convolution neural network (CNN) is further implemented in detecting and classifying the human behaviors using this information. Both the convolution layers and architecture of CNNs are presented. The analysis on loss and accuracy of training results are also shown. The experimental result indicates a precise determination of human motion behavior detection using the proposed system.

1 April 2019
Verified
Conference Paper
Closed access
FJFeng Jin
Feng Jin
RZRenyuan Zhang
Renyuan Zhang
ASArindam Sengupta
Arindam Sengupta
SCSiyang Cao
Siyang Cao
SHSalim Hariri
Salim Hariri
SASumit K. Agarwal
Sumit K. Agarwal
To address potential gaps noted in patient monitoring in the hospital, a novel patient behavior detection system using mmWave radar and deep convolution neural network (CNN), which supports the simultaneous recognition of multiple patients' behaviors in real-time, is proposed. In this study, we use an mmWave radar to track multiple patients and detect the scattering point cloud of each one. For each patient, the Doppler pattern of the point cloud over a time period is collected as the behavior signature. A three-layer CNN model is created to classify the behavior for each patient. The tracking and point clouds detection algorithm was also implemented on an mmWave radar hardware platform with an embedded graphics processing unit (GPU) board to collect Doppler pattern and run the CNN model. A training dataset of six types of behavior were collected, over a long duration, to train the model using Adam optimizer with an objective to minimize cross-entropy loss function. Lastly, the system was tested for real-time operation and obtained a very good inference accuracy when predicting each patient's behavior in a two-patient scenario.

19 September 2019 in IEEE Access
Verified
Journal Article
Open access
RZRenyuan Zhang
Renyuan Zhang
SCSiyang Cao
Siyang Cao
In this paper, a new radar-camera fusion system is presented. The fusion system takes consideration of error bounds of the two different coordinate systems from the inhomogeneous sensors, and further designed a new extended fusion-Kalman filter to adapt to the two sensors. The application details such as unsynchronization between sensors, multi-target tracking and association are also considered and illustrated in this paper. Experiment shows that the proposed fusion system can realize a range accuracy of 0.29m with a angular accuracy of 0.013rad in real-time. Therefore, the proposed fusion system is effective, reliable and computational efficient for real-time kinematic fusion applications.

1 November 2019
Verified
Conference Paper
Closed access
RZRenyuan Zhang
Renyuan Zhang
SCSiyang Cao
Siyang Cao
In this paper, a robust and adaptive radar point cloud clustering algorithm, named radar elliptical density-based spatial clustering of applications with noise (REDBSCAN), is presented. The proposed algorithm shows better clustering results for adapting to the arbitrary shape of targets as well as any number of targets comparing with traditional clustering methods. The algorithm is presented and is implemented in experiments using the state-of-art mmWave radar sensor with multiple-input multiple-output (MIMO) antennas. The related signal processing chain and the clustering outcomes are also discussed.

1 April 2018
Verified
Conference Paper
Closed access
RZRenyuan Zhang
Renyuan Zhang
SCSiyang Cao
Siyang Cao
This paper studies classification of the automotive radar interference waveforms via support vector machine (SVM). Automotive radar implemented in advanced driver assistance systems (ADASs) is an essential sensor in road traffic safety, e.g., moving target indication, collision avoidance and enhanced navigation system. However, radar-to-radar interference is inevitable as the number of automotive radar increases. Our work shows different types of radar-to-radar interference with analyzing the received signal. Providing linear frequency modulated transmitting (LFM) signal, filtering and dechirping techniques, the classification of six different types of radar-to-radar interference are presented and analyzed. Time-frequency domain signal and range-Doppler profiles of different types of interference are simulated. The machine learning classifier of SVM of multi-class high-dimensional waveform data classification is used to classify different interference waveforms and non-interference waveforms. Random produced dataset is cross validated through the SVM classifier. Different types of interference prediction accuracies are shown and verified by the proposed method. The confusion matrix of interference and non-interference, detecting different types of interference and classifying all incoming receiving signal are presented.