Motion States Classification of Rotor Target Based On Micro-Doppler Features Using CNN

Wang, Wantian; Tang, Ziyue; Xiong, Xin; Chen, Yi‐Chang; Zhang, Yuanpeng; Sun, Yongjian; Zhu, Zhenbo; Zhou, Chang

doi:10.1109/igarss.2019.8900361

Cited by 4 publications

(2 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There are works using the property of radar to detect velocity to identify the micro-Doppler signatures of drone rotors, e.g., [35]- [39]. CNNs are also applied for the classification of drones or their motion states based on micro-Doppler signatures in, e.g., [40], [41]. For detection and classification, a deep CNN is applied in [42] and a long short-term memory neural network for detection, classification, and localization is utilized in [43], in which they handle the localization by determining the angle for the received micro-Doppler pattern over time.…”

Section: B Related Workmentioning

confidence: 99%

Indoor Localization based on Short-Range Radar and Rotating Landmarks

Thormann,

Steuernagel,

Baum

2024

Preprint

View full text Add to dashboard Cite

A novel concept for indoor self-localization based on relative position measurements to rotating artificial landmarks (with known positions) using short-range radar is proposed. This includes a complete processing pipeline for extracting distance and angle measurements from the raw radar data, which consists of a neural network for distance estimation, a basic angle-of-arrival estimator, and a particle filter for position tracking. Due to the ability of radar to measure range rate, i.e., the velocity in the direction of a detection, it is possible to robustly detect the landmarks by detecting and localizing their micro-Doppler pattern. This mean localization is possible even under difficult conditions (e.g., light changes). Experiments with a wheeled mobile robot and common office fans as landmarks demonstrate the effectiveness of the approach for indoor localization.

show abstract

Section: B Related Workmentioning

confidence: 99%

Indoor Localization based on Short-Range Radar and Rotating Landmarks

Thormann,

Steuernagel,

Baum

2024

Preprint

View full text Add to dashboard Cite

show abstract

“…The CNN classification is used to determine the micro-motion category of the target. In a similar approach, W. Wang et al [53] first computed the time-frequency maps that reflect the micro-Doppler of the target, and then employ CNN to classify the three motion states of the rotor target: hover, ascent, and descent. The existing research on JEM classification is limited, with most classifications being based on time-frequency transformation.…”

Section: Jem Signal Based Radar Object Classificationmentioning

confidence: 99%

A Mutual-Attention Guided Feature Extraction and Adaptative Decision Fusion Framework for Fine-Grained Dual-Band Radar Target Classification

Zhong,

Chen,

Wang

et al. 2024

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

Deep Learning-Based Drone Classification Using Radar Cross Section Signatures at mmWave Frequencies

Al-Absi

Kim

et al. 2021

IEEE Access

View full text Add to dashboard Cite

This research was a part of the project titled 'Marine digital AtoN information management and service system development(1/5) (20210650)', funded by the Ministry of Oceans and Fisheries, Korea.."ABSTRACT This paper presents drone classification at millimeter-wave (mmWave) radars using the deep learning (DL) technique. The adoption mmWave technology in radar systems enables better resolution and aid in detecting smaller drones. Using radar cross-section (RCS) signature enables us to detect malicious drones and suitable action can be taken by respective authorities. Existing drone classification converts the RCS signature into images and then performs drone classification using a convolution neural network (CNN).Converting every signature into an image induces additional computation overhead; further CNN model is trained considering fixed learning rate. Thus, when using CNN-based drone classification under a highly dynamic environment exhibit poor classification accuracy. This paper present an im-proved long short-term memory (LSTM) by introducing a weight optimization model that can reduce computation overhead by not allowing the gradient to not flow through hidden states of the LSTM model. Further, present adaptive learning rate optimizing (ALRO) model for training the LSTM model. Experiment outcome shows LSTM-ALRO achieves much better drone detection accuracies when compared with the existing CNN-based drone classification model. INDEX TERMSConvolutional neural network, drone detection, micro doppler signature (MDS), millimeter-wave, radar cross-section, unmanned aerial vehicle.

show abstract

Motion States Classification of Rotor Target Based On Micro-Doppler Features Using CNN

Cited by 4 publications

References 12 publications

Indoor Localization based on Short-Range Radar and Rotating Landmarks

Indoor Localization based on Short-Range Radar and Rotating Landmarks

A Mutual-Attention Guided Feature Extraction and Adaptative Decision Fusion Framework for Fine-Grained Dual-Band Radar Target Classification

Deep Learning-Based Drone Classification Using Radar Cross Section Signatures at mmWave Frequencies

Contact Info

Product

Resources

About