Application of Auxiliary Classifier Wasserstein Generative Adversarial Networks in Wireless Signal Classification of Illegal Unmanned Aerial Vehicles

Abstract

Recently, many studies have reported on image synthesis based on Generative Adversarial Networks (GAN). However, the use of GAN does not provide much attention on the signal classification problem. In the context of using wireless signals to classify illegal Unmanned Aerial Vehicles (UAVs), this paper explores the feasibility of using GAN to improve the training datasets and obtain a better classification model, thereby improving the accuracy of classification. First, we use the generative model of GAN to generate a large datasets, which does not need manual annotation. At the same time, the discriminative model of GAN is improved to classify the types of signals based on the loss function of the discriminative model. Finally, this model can be used to the outdoor environment and obtain a real-time illegal UAVs signal classification system. Our experiments confirmed that the improvements on the Auxiliary Classifier Generative Adversarial Networks (AC-GANs) by limited datasets achieve excellent results. The recognition rate can reach more than 95% in the indoor environment, and this method is also applicable in the outdoor environment. Moreover, based on the theory of Wasserstein GANs (WGAN) and AC-GANs, a more robust Auxiliary Classifier Wasserstein GANs (AC-WGANs) model is obtained, which is suitable for multi-class UAVs. Through the combination of AC-WGANs and Universal Software Radio Peripheral (USRP) B210 software defined radio (SDR) platform, a real-time UAVs signal classification system is also implemented.

Orchestrating the Development Lifecycle of Machine Learning-based IoT Applications

“…AE can be used for diagnosis/fault detection tasks [65,250] or simply as a preprocessing tool (i.e., feature extraction/dimensionality reduction). GAN has been used for studies on generating rare category samples, and this upsampling approach may further improve the model performance [390,391].…”

“…Moreover, GAN has been used as a tool to generate large datasets that do not need manual annotation. For example, a study [390] explored the feasibility of using GAN to generate illegal Unmanned Aerial Vehicles (UAVs) dataset and obtain a better classification model with better accuracy. As the data generated from sensors may be unlabelled, GANs may have more potential applications in the IoT environment.…”

Recognition of UAV Video Signal Using RF Fingerprints in the Presence of WiFi Interference

“…Most UAVs have three RF signal types: remote control, telemetry, and video signal (VS) [9][10]. The first one is a frequency hopping uplink signal to control the UAV, while the other two are downlink signals.…”

“…Nevertheless, this strategy is sensitive to ambient noise. The work of [10] studies the WiFi and UAV VS envelope as eigenvalues, and the auxiliary classifier general adverse networks (AC-WGANs) model is used to recognize them. In theory, these methods can distinguish well UAV VS from WiFi.…”

A Size-Controlled AFGAN Model for Ship Acoustic Fault Expansion

“…The GAN approach has been applied to image recognition [12], audio [13], video [14], and other fields [15][16][17][18]. However, only a small amount of work has been reported related to fault vibration signal identification [19].…”