A5016387492- Target Tracking and Data Fusion in Sensor Networks
- Advanced Wireless Communication Techniques
- Contact Mechanics and Variational Inequalities
- Advanced Measurement and Metrology Techniques
- Elasticity and Material Modeling
- Stability and Control of Uncertain Systems
- Advanced Algorithms and Applications
- Parallel Computing and Optimization Techniques
- Video Surveillance and Tracking Methods
- Stability and Controllability of Differential Equations
- Reservoir Engineering and Simulation Methods
- Software System Performance and Reliability
- Neural dynamics and brain function
- Control Systems and Identification
- Neural Networks and Applications
- Cloud Computing and Resource Management
- Advanced Numerical Analysis Techniques
- Human Mobility and Location-Based Analysis
- Manufacturing Process and Optimization
- Atmospheric and Environmental Gas Dynamics
- Wireless Communication Networks Research
- Anomaly Detection Techniques and Applications
- Remote Sensing and Land Use
- Seismic Imaging and Inversion Techniques
- Neural Networks and Reservoir Computing
Zhejiang University of Science and Technology
2024
Dalian University of Technology
2004-2013
IBM Research (China)
2011
Zhejiang University
2011
Smart cities have drawn a lot of interest in recent years, which employ Internet Things (IoT)-enabled sensors to gather data from various sources and help enhance the quality residents’ life multiple areas, e.g. public safety. Accurate crime prediction is significant for safety promotion. However, complicated spatial-temporal dependencies make task challenging, due two aspects: 1) spatial dependency includes correlations with spatially adjacent regions underlying distant regions, mobility...
The paper focuses on the Kalman filtering problem for linear continuous-time systems with multiple delayed measurements. An explicit and simpler solution to is presented such systems. approach applied reorganised innovation analysis. obtained filter given in terms of Riccati differential equations. A numerical example demonstrate proposed approach.
This note studies the finite horizon H/sub /spl infin// fixed-lag smoothing problem for linear continuous time-varying systems. A technique named as reorganized innovation analysis in Krein space is developed to give a necessary and sufficient condition existence of an smoother. The given terms boundedness two matrix functions which are derived from solutions Riccati differential equations (RDEs), one standard filtering RDE 2/ type RDE. Examples demonstrate proposed smoother design fact that...
Gaussian particle filter algorithm provides a framework to estimate the state of moving object. However, it is known fact that parameters like noise variance and number affect effectiveness greatly. To improve performance in contour tracking, authors propose parameter adapting mechanism. simplify filter's implementation, variant sampling method also proposed. This combines step with prediction by taking advantage assumption exploring linear structure system dynamic model. Finally,...
We consider a cloud data center in which set of application servers is hosted. Each server runs virtual machine the cloud, subjected to session-oriented workload, whereby session one replicated on other for purposes high availability. In such an environment, we are concerned with memory resource usage due servers. particular, our goal prevent overload by managing load admitted each Little interest has been given management based usage, although crucial and valuable resource. introduce...
In this paper, a novel Gaussian process echo state networks (GPESN) model is proposed for time series forecasting. This method establishes the direct relationship between prediction origin and horizon without iterating in process, which avoids accumulative iteration error. Instead of using linear regression, used to obtain reservoir network output ESN, eliminates ill conditioned matrix. The GPESN capable achieving not only better result but also an accurate probability estimation results....