A5041907769- Caching and Content Delivery
- Opportunistic and Delay-Tolerant Networks
- IPv6, Mobility, Handover, Networks, Security
- Transportation Planning and Optimization
- Indoor and Outdoor Localization Technologies
- Traffic Prediction and Management Techniques
- Internet Traffic Analysis and Secure E-voting
- Underwater Vehicles and Communication Systems
- Artificial Immune Systems Applications
- Robotic Path Planning Algorithms
- Traffic and Road Safety
- Network Security and Intrusion Detection
- Cooperative Communication and Network Coding
- Urban Transport and Accessibility
- Traffic control and management
- Peer-to-Peer Network Technologies
- Mobile Ad Hoc Networks
- Psychiatry, Mental Health, Neuroscience
- Software-Defined Networks and 5G
- Urban Heat Island Mitigation
- Robotics and Sensor-Based Localization
- Remote Sensing and LiDAR Applications
- Smart Materials for Construction
- Emotion and Mood Recognition
- Wildlife-Road Interactions and Conservation
Fudan University
2020-2024
Changsha University of Science and Technology
2019-2020
Durham University
2018
Accurate traffic speed prediction is an important and challenging topic for transportation planning. Previous studies on predominately used spatio-temporal context features prediction. However, they have not made good use of the impact incidents. In this work, we aim to make information incidents achieve a better speed. Our incident-driven framework consists three processes. First, propose critical incident discovery method discover with high Second, design binary classifier, which uses deep...
Affective Behavior Analysis aims to facilitate technology emotionally smart, creating a world where devices can understand and react our emotions as humans do. To comprehensively evaluate the authenticity applicability of emotional behavior analysis techniques in natural environments, 6th competition on in-the-wild (ABAW) utilizes Aff-Wild2, Hume-Vidmimic2, C-EXPR-DB datasets set up five competitive tracks, i.e., Valence-Arousal (VA) Estimation, Expression (EXPR) Recognition, Action Unit...
The presence of raindrop induced image distortion has a significant negative impact on the performance wide range all-weather visual sensing applications including within increasingly important contexts surveillance and vehicle autonomy. A key part this problem is robust detection such that potential for degradation in effected regions can be identified. Here we address colour video imagery by considering three varying region proposal approaches with secondary classification via number novel...
Accurate traffic speed prediction is an important and challenging topic for transportation planning. Previous studies on predominately used spatio-temporal context features prediction. However, they have not made good use of the impact urban incidents. In this work, we aim to make information incidents achieve a better speed. Our incident-driven framework consists three processes. First, propose critical incident discovery method discover with high Second, design binary classifier, which...
A challenging rescue task for the underground disaster is to guide survivors in getting away from dangerous area quickly. To address issue, an escape guidance path developing method proposed based on anisotropic wireless sensor networks under condition of sparse anchor nodes. Firstly, a hybrid channel model was constructed reflect relationship between distance and receiving signal strength, which incorporates complex communication characteristics, including analytical ray wave model,...
CDN is a critical Internet infrastructure that helps users get contents with short delay. With the development of application scenarios, requests will involve more than one resource type. Unfortunately, existing server selection schemes targeting type cannot select most suitable by considering different types together. In this paper, we propose Polygon, system supporting multiple demands. The keystone Polygon deployment set dispatchers at strategic network locations, which can be accessed...
Today, Internet service deployment is typically implemented with server replication at multiple locations. Domain name system (DNS), which translates human-readable domain names into network-routable IP addresses, used for distributing users to different replicas. However, DNS relies on several network-based queries and the delay connection setup process between client replica. In this article, we propose Artemis, a practical low-latency naming routing that supports optimal (replica)...
Understanding the network usage patterns of university users is very important today. This paper focuses on research DNS request behaviors in Shanghai, China. Based logs a large number recorded by CERNET, we conduct general analysis behavior browsing from two perspectives: characteristics users’ and market share CDN service providers. We also undertake experiments requests for providers using different prediction models. Firstly, order to understand Internet access patterns, select top seven...
Determining the physical locations of CDN Points Presence (PoPs) is fundamental to understanding and diagnosing services. Yet, popular deployment IP Anycast in CDNs has rendered existing geolocation tools unreliable. To fill this gap, we present an HTTP-based solution that leverages subtle geographic hints HTTP responses locate PoPs at city-level granularity. The evaluation shows our technique achieves over 90% accuracy with average error distance less than 40 km.
Online Social Networks (OSNs) play a significant role in people's daily life. Increasing OSN traffic promotes the requirement for building self-operated Content Delivery (CDNs) to deliver media data efficiently and reduce costs. CDNs is heavily influenced by social connectivity, such as friendships. To CDN network better using connectivity information, we propose SocialCache, pervasive social-aware caching strategy CDNs. SocialCache supports several metric options that reflect importance of...
A challenging rescue task for the underground disaster is to guide survivors getting away dangerous area quickly. To address issue, an escape guidance path developing method proposed based on deployed wireless sensor network. Firstly, a hybrid channel model constructed incorporate complex communication characteristics. Then anchors which have valid location information are utilized locate other network nodes. Under sparse conditions after disasters, anchor chosen algorithm select has...