A5089095312- Advanced Optical Network Technologies
- Advanced Photonic Communication Systems
- Video Surveillance and Tracking Methods
- Optical Network Technologies
- Neural Networks and Reservoir Computing
- Environmental and Agricultural Sciences
- Anomaly Detection Techniques and Applications
- Advanced Neural Network Applications
- Remote Sensing in Agriculture
- IoT-based Smart Home Systems
- Gait Recognition and Analysis
- Advanced Memory and Neural Computing
- Digital Marketing and Social Media
- Remote Sensing and Land Use
- IoT and Edge/Fog Computing
- Privacy-Preserving Technologies in Data
- Technology Adoption and User Behaviour
- Optical Wireless Communication Technologies
- Consumer Retail Behavior Studies
- Software-Defined Networks and 5G
Beijing University of Posts and Telecommunications
2019-2024
Nanyang Technological University
2022
The Industrial Internet of Things (IIoT) accommodates a huge number heterogeneous devices to bring vast services under distributed computing scenarios. Most productive in IIoT are closely related production control and require network support with low delay. However, the resource reservation based on gross traffic prediction ignores importance treats them as ordinary services, so it is difficult provide stable delay for large amounts service requests. For many productions, unexpected...
Variations in sales promotion e-commerce impact online purchase decisions. Managers of marketplaces must comprehend the effects deviations on We categorize changes as either recognisable or unrecognisable to consumer. used stimulus-organization-response model assess influence diverse promotions consumer decisions via perceived attractiveness product. Using data from 587 consumers we discovered distinct impacts between and intention. Additionally, product value moderates these effects....
With a focus on practical applications in the real world, number of challenges impede progress pedestrian detection. Scale variance, cluttered backgrounds and ambiguous features are main culprits detection failures. According to existing studies, consistent feature fusion, semantic context mining inherent attributes seem be feasible solutions. In this paper, tackle prevalent problems detection, we propose an anchor-free detector, named attribute perception (CAPNet). particular, first...
Precise location of personnel indoors is one the crucial prerequisite data analytics workforce productivity as well workplace safety and health. In this paper, we propose an enhanced indoor occupancy tracking system using optical camera communication (OCC) on top conventional surveillance cam-eras. The proposed able to track workers who carry unique infrared LED beacons indoor. OCC with adopted for identity check, while localization accuracy deep-learning-based human pose estimation infer...
We demonstrate a novel intent defined optical network (IDON) platform that introduces self-adapted generation and optimization (SAGO) policy, utilizing closed-loop policy guarantee to achieve zero-touch operation of network.
This paper proposes a brain-like development based multi-routing combination optimization scheme for high mobility in optical fronthaul. The experimental results show that the can significantly reduce delay, communication blocking rate, and improve quality.
This paper proposes a brain-like resource allocation strategy for data center optical networks with high bandwidth demand differences. Experiments show that this can improve utilization, reduce network latency and blocking rate.
Abstract China is the world’s largest rice producer and consumer. Accurate identification of planting irrigation area great significance to China’s food security. Based on GF-1 multispectral data, spectral matching OTSU algorithm were used monitor in Chuanhang Irrigation Area from 2016 2020. The accuracy was verified by confusion matrix. overall 98.1691%, 91.8706%, 95.2451%, 93.0604%, 91.9697%, respectively. Kappa coefficients 0.9541, 0.7898, 0.8777, 0.8330 0.7824, respectively, which proved...