A5101998597- Neural Networks and Applications
- Robotic Mechanisms and Dynamics
- Adaptive Control of Nonlinear Systems
- Control Systems and Identification
- Model Reduction and Neural Networks
- Advanced Numerical Analysis Techniques
- Wireless Body Area Networks
- Advanced Algorithms and Applications
- Energy Harvesting in Wireless Networks
- Image and Video Stabilization
- Advanced Sensor and Control Systems
- Molecular Communication and Nanonetworks
- Industrial Technology and Control Systems
- Iterative Learning Control Systems
- Energy Efficient Wireless Sensor Networks
- Chaos control and synchronization
- Bluetooth and Wireless Communication Technologies
- Advanced Adaptive Filtering Techniques
- Inertial Sensor and Navigation
- Distributed Control Multi-Agent Systems
- Guidance and Control Systems
- Context-Aware Activity Recognition Systems
- Digital Imaging for Blood Diseases
- Brain Tumor Detection and Classification
- Digital Filter Design and Implementation
Guangdong Polytechnic Normal University
2022-2024
Jiangxi University of Science and Technology
2011-2020
Gannan Normal University
2019
Hong Kong Polytechnic University
2018
Shenzhen Institutes of Advanced Technology
2013-2017
Chinese Academy of Sciences
2013-2017
Shenzhen University
2017
Shenzhen Institute of Information Technology
2013
Sun Yat-sen University
2008-2010
Energy efficiency is a key concern for wireless sensor nodes, especially body area network (WBAN) in which sensors operate close vicinity to, on or even inside human body. In this paper, we first present system-level energy consumption model associated with transmission distance d and data rate over on-body communication link. Then, based the analysis of tradeoff between circuit distance, threshold <sub xmlns:mml="http://www.w3.org/1998/Math/MathML"...
A new recurrent neural network (RNN) is presented for solving online linear time-invariant (LTI) equations, which has been developed based ingeniously on a vector-valued error-function rather than scalar-valued norm-based function. Theoretical analysis and simulation results both substantiate the efficacy of such an RNN model LTI equation solving.
Based on polynomial interpolation and approximation theory, a special feed-forward neural network using power activation functions is constructed in this paper. The model employs three-layer structure with the hidden-layer neurons activated by group of order-increasing (while other layers' use linear functions). In addition, weights-updating formula for such could be derived from standard BP training method. A pseudoinverse-based method (or termed, weights-direct-determination /...
The core problem with nodes of wireless sensor networks is the limited charge capacity for power-operated battery. main concern this paper how to extend battery's lifetime. According characteristics and functionality nodes, applications are categorized into three distinct types, i.e., hard real-time applications, soft periodic applications. For a battery-friendly lazy packet scheme designed draw smoother lower current minimizing battery consumption. A local optimization slack time exploited...
It is well-known that, the classical gradient-descent-based neural network (CGNN) model used widely for time-invariant problem solving. However, it an extremely common time varying cases in practical engineering applications, while CGNN effective solver time-variant problems. For this reason, article, adaptive GNN (AGNN) presented linear variant matrix equation (LTVME) on-line solving based on Lyapunov theory. Theoretical analysis already verified that AGNN could achieve correct state...
This paper focuses on the problem of target tracking using k fittest robots in a group n mobile with > . We present centralized and distributed coordination models all‐to‐all limited communications, respectively. For case communication between robots, theoretical analysis is presented to prove exponential stability whole system. In real applications robotic networks, robot may only be allowed exchange information number neighbors. such where quantity not available, consensus filter used...
For satisfying requirements of the medical application, IEEE 802.15 Task Group approved 802.15.6 communication standard which is optimized for low power device. In this paper, we analyze performance MAC protocol, in terms throughput, delay-time and energy consumption under saturation unsaturation conditions. Considering idle state nodes, presented a new three-dimensional Markov model to evaluate Carrier Sense Multiple Access with Collision Avoidance (CSMA/CA) mechanism based on protocol. The...
Since 12 March 2001, Zhang et al have proposed a special class of recurrent neural networks for online time-varying problems solving, especially matrix inversion. For possible hardware (e.g., digital-circuit) realization, such (ZNN) could also be reformulated in the discrete-time form, which incorporates Newton iteration as case. In this paper, constant inversion, we generalize and investigate more ZNN models (which termed iterations) by using multiple-point backward-difference formulas....
Energy consumption minimization is a key concern for wireless sensor nodes, especially body area network (WBAN) in which sensors operate close vicinity to, on or even inside human body. In this paper, system level energy model with transmission distance and rate on-body communication link addressed. Based analysis of the trade-off between circuit distance, threshold responsible proportion derived. Then different optimal schemes needed various are presented; finally, an illustrative example...