Search

The development of civil aviation has led to flight operations generating massive datasets. Derived from Automatic Dependent Surveillance-Broadcast technology, an eigengap-based automatic hierarchical clustering algorithm is proposed, aiming to overcome the human intervention requirement of hierarchical clustering to determine the number of track clusters in the terminal area. First, the trajectory pair similarity matrix is calculated based on the Euclidean distance and an adaptive scale parameter. Second, the Laplace transformation is applied on the similarity matrix and the eigenvalues are obtained in order to determine the number of clusters. Finally, the hierarchical clustering algorithm divides the terminal area trajectory into several sub-categories. Taking a terminal approach trajectory as an example, simulation analysis is performed, with results revealing that the algorithm divides 404 north-south-oriented aircraft flight trajectories into 2 and 3 categories. Furthermore, the new-index is used to evaluate the clustering results, demonstrating its effectiveness is better than the automatic k-means algorithm. Our study provides support for reliability analysis of air traffic network in the terminal area.

iWith the rapid development and the increasing security threats of D2D communication, the D2D communication security has become more and more important. Device recognition is the foundation of communication security. Traditional device authentication is the cryptographic mechanism and the security protocol that are not enough for the new security threat. RF fingerprint of wireless device has physical characteristics, which is difficult to clone. Therefore, it can be used for D2D device recognition. In this paper, a recognition algorithm of D2D device based on RF fingerprint is proposed. First, Hilbert transform and principal component analysis is used to generate the RF fingerprint of D2D device. Second, support vector machine (SVM) is introduced and CV-SVM is used as the classifier. Finally, simulation results show that the recognition rate is more than 90% under SNR = 5 dB, which proves that the proposed method is effective for D2D device recognition. Therefore, it can be used as the trust management of D2D communication.

Automated modulation classification plays a very important part in cognitive radio networks. Deep learning is also a powerful tool that we could not overlook its potential in addressing signal modulation recognition problem. In our last work, we propose a new data conversion algorithm in order to gain a better classification accuracy of communication signal modulation, but we still believe that the convolution neural network (CNN) can work better. However, its application to signal modulation recognition is often hampered by insufficient data and overfitting. Here, we propose a smart approach to programmatic data augmentation method by using the auxiliary classifier generative adversarial networks (ACGANs). The famous CNN model, AlexNet, has been utilized to be the classifier and ACGAN to be the generator, which will enlarge our data set. In order to alleviate the common issues in the traditional generative adversarial nets training, such as discriminator overfitting, generator disconverge, and mode collapse, we apply several training tricks in our training. With the result on original data set as our baseline, we will evaluate our result on enlarged data set to validate the ACGAN’s performance. The result shows that we can gain 0.1~6% increase in the classification accuracy in the ACGAN-based data set.