Publications by authors named "Jingtang Hao"

With the growing integration of drones into various civilian applications, the demand for effective automatic drone identification (ADI) technology has become essential to monitor malicious drone flights and mitigate potential threats. While numerous convolutional neural network (CNN)-based methods have been proposed for ADI tasks, the inherent local connectivity of the convolution operator in CNN models severely constrains RF signal identification performance. In this paper, we propose an innovative hybrid transformer model featuring a CNN-based tokenization method that is capable of generating T-F tokens enriched with significant local context information, and complemented by an efficient gated self-attention mechanism to capture global time/frequency correlations among these T-F tokens.

View Article and Find Full Text PDF
Article Synopsis
  • * The proposed solution uses a gated recurrent unit (GRU)-based interacting multiple model (IMM) approach to enhance tracking efficiency by creating various motion models and dynamically selecting the best one for different tracking scenarios.
  • * Simulation results on bird trajectory data indicate that this new method reduces tracking errors by at least 26% compared to traditional tracking methods, demonstrating its effectiveness for Lidar systems and other low-refresh-rate tracking applications.
View Article and Find Full Text PDF

The localization accuracy is susceptible to the received signal strength indication (RSSI) fluctuations for RSSI-based wireless localization methods. Moreover, the maximum likelihood estimation (MLE) of the target location is nonconvex, and locating target presents a significant computational complexity. In this paper, an RSSI-based access point cluster localization (APCL) method is proposed for locating a moving target.

View Article and Find Full Text PDF