Signal Preprocessing Technique With Noise-Tolerant for RF-Based UAV Signal Classification

Abstract

Since the beginning of the COVID-19 pandemic, the demand for unmanned aerial vehicles (UAVs) has surged owing to an increasing requirement of remote, noncontact, and technologically advanced interactions. However, with the increased demand for drones across a wide range of fields, their malicious use has also increased. Therefore, an anti-UAV system is required to detect unauthorized drone use. In this study, we propose a radio frequency (RF) based solution that uses 15 drone controller signals. The proposed method can solve the problems associated with the RF based detection method, which has poor classification accuracy when the distance between the controller and antenna increases or the signal-to-noise ratio (SNR) decreases owing to the presence of a large amount of noise. For the experiment, we changed the SNR of the controller signal by adding white Gaussian noise to SNRs of -15 to 15 dB at 5 dB intervals. A power-based spectrogram image with an applied threshold value was used for convolution neural network training. The proposed model achieved 98% accuracy at an SNR of -15 dB and 99.17% accuracy in the classification of 105 classes with 15 drone controllers within 7 SNR regions. From these results, it was confirmed that the proposed method is both noise-tolerant and scalable. INDEX TERMSAnti-drone systems, convolutional neural networks (CNNs), noise-tolerant, spectrograms, unmanned aerial vehicles (UAVs), UAV classification I. INTRODUCTION U NMANNED aerial vehicles (UAVs), including drones, are used for various purposes, such as delivery, agriculture, transportation, and communication [1]. The potential uses of such vehicles are continually increasing [2]. Additionally, social and commercial demands for remote technology have increased owing to the recent COVID-19 outbreak, and drones are being proposed as a noncontact solution for numerous applications. The approach in [3] shows how backup transportation systems based on existing drone infrastructure can play an important role during the COVID-19 pandemic and similar situations. Additionally, the authors in [4] proposed the application of drones for spraying disinfectants to combat the COVID-19 pandemic. However,

Malicious UAV Detection Over Rician Fading Channel: Performance Analysis

“…Experimental results in [19] showed that the proposed hierarchical learning approach has outperformed other techniques with an accuracy rate of 0.99. Convolution neural network (CNN) has been utilized in [20] to develop a noise immune UAV signal classification system which has been tested at different signal-tonoise ratios (SNRs). Support vector machine (SVM), artificial neural network (ANN), decision tree (DT), and random forest (RandF) classifiers have been utilized in [21] for detection and classification of drones in the presence of Wi-Fi interference using UAV video signal (VS) fingerprints where indoor and outdoor experiments have been carried out with high accuracy rates, especially by using RandF.…”

“…where ϵ is the weighting factor that represent the contribution of H 1 on Ψ such that 0 < ϵ < 1. Therefore, the expression of the suboptimal threshold under the LoS channel state, η L , can be expressed using (20) as…”

“…. , 1] with a proper κ value 2: define φ = 0.5, ε = 0 and i = 1 as initial values 3: evaluate Ψ at ϵ(i) from (20) 4: evaluate η L and ϕ from ( 21) and ( 22) 5: while ϕ < φ do 6: ability of detection in low-complexity suboptimal receiver using the derived thresholds is…”

Intelligent Identification and Classification of Small UAV Remote Control Signals Based on Improved Yolov5-7.0

“…It can be seen from the relevant studies that all types of studies show a relatively complete research system, and the relevant recent studies have been organized in detail in [28].…”

RF Drone Detection System Based on a Distributed Sensor Grid With Remote Hardware-Accelerated Signal Processing

“…Although the proposed approach is based on timedomain 250 µs signal segments with minor input signal processing, it has the potential to achieve high accuracy and be used in portable real-time systems while having a relatively low level of complexity. In contrast, to increase accuracy in low SNR conditions, special preprocessing techniques are applied to the time-frequency signal representation in [33] prior to utilising DL. The method's findings imply that distinguishing using the spectrogram is more accurate, however, the classification of multiple drone controller signals remains challenging.…”