Abstract
With the upsurge in the use of Unmanned Aerial Vehicles (UAVs) in various fields, detecting and identifying them in real-time are becoming important topics. However, the identification of UAVs is difficult due to their characteristics such as low altitude, slow speed, and small radar cross-section (LSS). With the existing deterministic approach, the algorithm becomes complex and requires a large number of computations, making it unsuitable for real-time systems. Hence, effective alternatives enabling real-time identification of these new threats are needed. Deep learning-based classification models learn features from data by themselves and have shown outstanding performance in computer vision tasks. In this paper, we propose a deep learning-based classification model that learns the micro-Doppler signatures (MDS) of targets represented on radar spectrogram images. To enable this, first, we recorded five LSS targets (three types of UAVs and two different types of human activities) with a frequency modulated continuous wave (FMCW) radar in various scenarios. Then, we converted signals into spectrograms in the form of images by Short time Fourier transform (STFT). After the data refinement and augmentation, we made our own radar spectrogram dataset. Secondly, we analyzed characteristics of the radar spectrogram dataset with the ResNet-18 model and designed the ResNet-SP model with less computation, higher accuracy and stability based on the ResNet-18 model. The results show that the proposed ResNet-SP has a training time of 242 s and an accuracy of 83.39%, which is superior to the ResNet-18 that takes 640 s for training with an accuracy of 79.88%.
Highlights
IntroductionThe rapid development of Unmanned Aerial Vehicle (UAV) technology has increased the usage of Unmanned Aerial Vehicles (UAVs) in various fields such as agriculture, industry, and military
In recent years, the rapid development of Unmanned Aerial Vehicle (UAV) technology has increased the usage of Unmanned Aerial Vehicles (UAVs) in various fields such as agriculture, industry, and military.Even though the use of UAVs brings convenience to life, it poses severe threats if abused by enemies or terrorists
The rapid development of Unmanned Aerial Vehicle (UAV) technology has increased the usage of UAVs in various fields such as agriculture, industry, and military
Summary
The rapid development of Unmanned Aerial Vehicle (UAV) technology has increased the usage of UAVs in various fields such as agriculture, industry, and military. A moving target generates the micro-Doppler effect by partial movements like the pendulum, rotation and vibration along with a constant Doppler shift induced from the main body This MDS is a unique characteristic of the target and is visually represented well in a spectrogram of the STFT of the radar signal. This spectrogram can be trained with a deep learning image classification model. We generated a radar spectrogram dataset with a variety of UAV flight dynamics and designed a lightweight deep learning classification model that learns the MDS of targets, suitable for the real-time system. We can see that each UAV spectrogram appears differently and using this characteristic, we can further perform deep learning-based UAV classification
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have