A5022475247- Wireless Signal Modulation Classification
- Advanced Graph Neural Networks
- Speech Recognition and Synthesis
- Robotics and Automated Systems
- Advanced SAR Imaging Techniques
- Hand Gesture Recognition Systems
- Big Data and Digital Economy
- Biometric Identification and Security
- Fractal and DNA sequence analysis
- Synthetic Aperture Radar (SAR) Applications and Techniques
- User Authentication and Security Systems
- Infrared Target Detection Methodologies
- DNA and Biological Computing
National University of Defense Technology
2022-2025
Beijing Jiaotong University
2019
Semantic communication is a promising technology used to overcome the challenges of large bandwidth and power requirements caused by data explosion. representation an important issue in semantic communication. The knowledge graph, powered deep learning, can improve accuracy while removing ambiguity. Therefore, we propose system based on graph. Specifically, our system, transmitted sentences are converted into triplets using Triplets be viewed as basic symbols for extraction restoration...
Feature point detection in inverse synthetic aperture radar (ISAR) images of space targets is the foundation for tasks such as analyzing target motion intent and predicting on-orbit status. Traditional feature methods perform poorly when confronted with low texture uneven brightness characteristics ISAR images. Due to nonlinear mapping capabilities, neural networks can effectively learn features from targets, providing new ideas detection. However, scarcity labeled image data presents a...
Context information is significant for semantic extraction and recovery of messages in communication. However, context not fully utilized the existing communication systems since relationships between sentences are often ignored. In this paper, we propose an extended context-based (ECSC) system text transmission, which within explored representation recovery. At encoder, self-attention segment-level relative attention used to extract sentences, respectively. addition, a gate mechanism...
In the low signal-to-noise ratio region, a large number of bit errors occur, and it may exceed channel error correction capability receiver. Traditional communication system use technology automatic repeat-request to deal with this problem, which is time consuming waste resources. To enhance reliability system, we investigate reasoning decoding at semantic level instead grammar level. particular, propose model for text transmission, assisting be more robust in terrible environments. Based on...
With the popularity of smartphones, more and users are accustomed to storing sensitive information related personal privacy, which puts forward higher requirements for current authentication system. In this paper, an implicit solution based on user's keystroke behavior smartphone usage is proposed. Firstly, a data collected by three types built-in sensors in smartphone. After performing smoothing normalization processing, we extract temporal features time domain using statistical methods....