A5102962148- Target Tracking and Data Fusion in Sensor Networks
- Radar Systems and Signal Processing
- Distributed Sensor Networks and Detection Algorithms
- Data Stream Mining Techniques
- Environmental Engineering and Cultural Studies
- Applied Advanced Technologies
- Autonomous Vehicle Technology and Safety
- Hand Gesture Recognition Systems
- Image Processing Techniques and Applications
- Structural Health Monitoring Techniques
- Inertial Sensor and Navigation
- Geochemistry and Geologic Mapping
- Advanced Technologies and Applied Computing
- Network Security and Intrusion Detection
- Anomaly Detection Techniques and Applications
- Fault Detection and Control Systems
- Robotics and Sensor-Based Localization
Beihang University
2019-2024
In extant radar signal processing systems, detection and tracking are carried out independently, detected measurements utilized as inputs to the procedure. Therefore, performance is highly associated with accuracy, this may severely degrade when detections include a mass of false alarms missed-targets errors, especially in dense clutter or closely-spaced trajectories scenarios. To deal issue, paper proposes novel method for integrating multiple hypothesis tracker processing. Specifically,...
Conventional multi-sensor multi-target multi-Bernoulli (MS-MeMBer) filters are based on the assumption that each target produces at most one measurement per time step. However, this is not always reasonable in practice as an extended can generate multiple measurements step due to recent improvement sensor resolution. In case, a potential estimation bias may occur current MS-MeMBer filters. Therefore, novel filter and its Gaussian inverse Wishart mixture implementation given paper....
Data mining in real-time data streams is associated with multiple types of uncertainty, which often leads the respective categorizers to make erroneous predictions related presence or absence complex events. But recognizing abnormal events, even those that occur extremely rare cases, offers significant support decision-making systems. Therefore, there a need for robust recognition mechanisms will be able predict recognize when an event occurs on stream. Considering this need, paper presents...
Deep Neural Network (DNN) pruning has emerged as a key strategy to reduce model size, improve inference latency, and lower power consumption on DNN accelerators. Among various techniques, block output channel have shown significant potential in accelerating hardware performance. However, their accuracy often requires further improvement. In response this challenge, we introduce separate, dynamic differentiable (SMART) pruner. This pruner stands out by utilizing learnable probability mask for...
Millimetre wave (mmWave) radar is a non-intrusive privacy and relatively convenient inexpensive device, which has been demonstrated to be applicable in place of RGB cameras human indoor pose estimation tasks. However, mmWave relies on the collection reflected signals from target, containing information difficult fully applied. This long-standing hindrance improvement accuracy. To address this major challenge, paper introduces probability map guided multi-format feature fusion model,...
Adaptive Kalman filtering has become a widely used state estimation method when the system noise statistics are not available priori. In this article, we consider application of Stein variational gradient descent (SVGD) in adaptive for linear Gaussian state-space model with inaccurate process statistics. The proposed algorithm separately estimates and parameters filter SVGD online inference. Compared to existing state-of-the-art filters, can approximate better thus provides higher accuracy,...
Radiation processing technology has been more and widely used in light industry, chemical food preservation, medical care, other industries. Therefore, this paper designs a kind of wireless control system based on artificial intelligence γ radiation method, to strengthen the application equipment, monitoring can its surrounding environment. The experimental results show that C/Si atomic ratio oxidized SiC fiber air treated by scheme is about 1.75, while SiC(Be) electron irradiation close 1 : 1.
In extant radar signal processing systems, detection and tracking are carried out independently, detected measurements utilized as inputs to the procedure. Therefore, performance is highly associated with accuracy, this may severely degrade when detections include a mass of false alarms missed-targets errors, especially in dense clutter or closely-spaced trajectories scenarios. To deal issue, paper proposes novel method for integrating multiple hypothesis tracker processing. Specifically,...