- Image and Video Quality Assessment
- Cloud Computing and Resource Management
- Caching and Content Delivery
- Service-Oriented Architecture and Web Services
- Peer-to-Peer Network Technologies
- Software-Defined Networks and 5G
- Network Traffic and Congestion Control
- IoT and Edge/Fog Computing
- Multimedia Communication and Technology
- Distributed and Parallel Computing Systems
- Context-Aware Activity Recognition Systems
- Mobile Agent-Based Network Management
- Video Coding and Compression Technologies
- Advanced Optical Network Technologies
- Semantic Web and Ontologies
- Software System Performance and Reliability
- Network Security and Intrusion Detection
- Cloud Computing and Remote Desktop Technologies
- Business Process Modeling and Analysis
- Distributed systems and fault tolerance
- Parallel Computing and Optimization Techniques
- Interconnection Networks and Systems
- Electronic Health Records Systems
- IPv6, Mobility, Handover, Networks, Security
- Healthcare Technology and Patient Monitoring
Ghent University
2016-2025
Ghent University Hospital
2010-2023
Imec the Netherlands
2018-2023
COMSATS University Islamabad
2023
National Yang Ming Chiao Tung University
2023
Universidade do Vale do Itajaí
2023
IMEC
2006-2022
Conservatoire National des Arts et Métiers
2021-2022
University of Waterloo
2019-2022
Pohang University of Science and Technology
2019-2022
Network function virtualization (NFV) has drawn significant attention from both industry and academia as an important shift in telecommunication service provisioning. By decoupling network functions (NFs) the physical devices on which they run, NFV potential to lead reductions operating expenses (OPEX) capital (CAPEX) facilitate deployment of new services with increased agility faster time-to-value. The paradigm is still its infancy there a large spectrum opportunities for research community...
Network Functions Virtualization (NFV) is an upcoming paradigm where network functionality virtualized and split up into multiple building blocks that can be chained together to provide the required functionality. This approach increases flexibility scalability as these allocated reallocated at runtime depending on demand. The success of this depends existence performance algorithms determine where, how are instantiated. In paper, we present evaluate a formal model for resource allocation...
Scalable and effective exploration remains a key challenge in reinforcement learning (RL). While there are methods with optimality guarantees the setting of discrete state action spaces, these cannot be applied high-dimensional deep RL scenarios. As such, most contemporary relies on simple heuristics such as epsilon-greedy or adding Gaussian noise to controls. This paper introduces Variational Information Maximizing Exploration (VIME), an strategy based maximization information gain about...
Although mobile devices are gaining more and capabilities (i.e. CPU power, memory, connectivity, ...), they still fall short to execute complex rich media data analysis applications. Offloading the cloud is not always a solution, because of high WAN latencies, especially for applications with real-time constraints such as augmented reality. Therefore has be moved closer user in form cloudlets. Instead moving complete virtual machine from cloudlet, we propose fine grained cloudlet concept...
The electrification of the vehicle fleet will result in an additional load on power grid. Adequately dealing with such pluggable (hybrid) electrical vehicles (PHEV) forms part challenges and opportunities evolution towards Smart Grids. In this paper, we investigate potential benefits using control mechanisms, that could be offered by a Home Energy box, optimizing energy consumption stemming from PHEV charging residential use case. We present smart strategies based quadratic programming for...
Network function virtualization has received attention from both academia and industry as an important shift in the deployment of telecommunication networks services. It is being proposed a path towards cost efficiency, reduced time-to-markets, enhanced innovativeness service provisioning. However, efficiently running virtualized services not trivial as, among other initialization steps, it requires first mapping virtual onto physical networks, thereafter scheduling functions networks. This...
In HTTP Adaptive Streaming, video content is temporally divided into multiple segments, each encoded at several quality levels. The client can adapt the requested to network changes, generally resulting in a smoother playback. Unfortunately, live streaming solutions still often suffer from playout freezes and large end-to-end delay. By reducing segment duration, use smaller temporal buffer respond even faster changes. However, since segments are subsequently, this approach susceptible high...
Count-based exploration algorithms are known to perform near-optimally when used in conjunction with tabular reinforcement learning (RL) methods for solving small discrete Markov decision processes (MDPs). It is generally thought that count-based cannot be applied high-dimensional state spaces, since most states will only occur once. Recent deep RL strategies able deal continuous spaces through complex heuristics, often relying on optimism the face of uncertainty or intrinsic motivation. In...
With significant advances in holographic display technology, a plethora of interactive applications, such as tele-conferencing and tele-surgery, are well on their way to integrating technologies. However, hologram-based applications will place demands networking infrastructure, which not supported today. These include support for ultra-low delays, high bandwidth, the ability coordinate, synchronize, dynamically adapt multiple data streams. This article articulates these challenges points out...
To evaluate whether a real-time electronic alert system or "AKI sniffer," which is based on the RIFLE classification criteria (Risk, Injury and Failure), would have an impact therapeutic interventions acute kidney injury progression.Prospective intervention study.Surgical medical intensive care unit in tertiary hospital.A total of 951 patients having 1,079 admission episodes were admitted during study period (prealert control group: 227, 616, postalert 236).Three phases compared: A 1.5-month...
In this paper routing, planning of working capacity, rerouting, and spare capacity in wavelength division multiplexing (WDM) networks are investigated. Integer linear programming (ILP) simulated annealing (SA) used as solution techniques. A complex cost model is presented. The assignment optimized with respect to three restoration strategies. benefit conversion, the choice fiber line system, influence parameter values discussed, different strategies Wavelength conversion found be limited...
The evolution toward grid and cloud computing as observed for over a decennium illustrates the crucial role played by (optical) networks in supporting today's applications. In this paper, we start from an overview of challenging applications both academic (further referred to scientific), enterprise (business) nonprofessional user (consumer) domains. They pose novel challenges, calling efficient interworking IT resources, processing storage, well network that interconnects them provides...
Virtual Reality (VR) devices are becoming accessible to a large public, which is going increase the demand for 360° VR videos. videos often characterized by poor quality of experience, due high bandwidth required stream video. To overcome this issue, we spatially divide video into tiles, so that each temporal segment composed several spatial tiles. Only tiles belonging viewport, region watched user, streamed at highest quality. The other instead lower We also propose an algorithm predict...
Nowadays, the Internet of Things (IoT) continues to expand at enormous rates. Smart Cities powered by connected sensors promise transform public services from transportation environmental monitoring and healthcare improve citizen welfare. Furthermore, over last few years, Fog Computing has been introduced provide an answer massive growth heterogeneous devices network. Nevertheless, providing a proper resource scheduling for delay-sensitive data-intensive in environments is still key research...
HTTP Adaptive Streaming (HAS) is quickly becoming the de facto standard for video streaming services. In HAS, each temporally segmented and stored in different quality levels. Rate adaptation heuristics, deployed at player, allow most appropriate level to be dynamically requested, based on current network conditions. It has been shown that today’s heuristics underperform when multiple clients consume same time, due fairness issues among clients. Concretely, this means negatively influence...
Fog computing extends the cloud paradigm by placing resources close to edges of network deal with upcoming growth connected devices. Smart city applications, such as health monitoring and predictive maintenance, will introduce a new set stringent requirements, low latency, since can be requested on-demand simultaneously multiple devices at different locations. It is then necessary adapt existing technologies future needs design architectural concepts help meet these strict requirements. This...
HTTP Adaptive Streaming (HAS) is becoming the de facto standard for Over-The-Top (OTT)-based video streaming services such as YouTube and Netflix. By splitting a into multiple segments of couple seconds encoding each these at quality levels, HAS allows client to dynamically adapt requested during playout react network changes. However, state-of-the-art selection heuristics are deterministic tailored specific configurations. Therefore, they unable cope with vast range highly dynamic settings....
In this paper, we demonstrate and evaluate a method to perform real-time object detection on-board UAV using the state of art YOLOv2 algorithm running on an NVIDIA Jetson TX2, GPU platform targeted at power constrained mobile applications that use neural networks under hood. This, as result comparing several cutting edge algorithms. Multiple evaluations present provide insights help choose optimal configuration given certain frame rate accuracy requirements. We propose how setup can be used...
Network virtualisation is considerably gaining attention as a solution to ossification of the Internet. However, success network will depend in part on how efficiently virtual networks utilise substrate resources. In this paper, we propose machine learning-based approach resource management. We model decentralised system and introduce learning algorithm each node link, providing self-organization capabilities. multiagent that carries out management coordinated way. The task these agents use...
The increasing popularity of head-mounted devices and 360° video cameras allows content providers to offer virtual reality streaming over the Internet, using a relevant representation immersive combined with traditional techniques. While this approach user freely move her head, location is fixed by camera's position within scene. Recently, an increased interest has been shown for free movement scenes, referred as six degrees freedom. One way realize capturing objects through number...
Autism Spectrum Disorder (ASD) is characterized by social interaction difficulties and communication difficulties. Moreover, children with ASD often suffer from other co-morbidities, such as anxiety depression. Finding appropriate treatment can be difficult symptoms of co-morbidities overlap. Due to these challenges, parents higher levels stress. This research aims investigate the feasibility empowering their through use a serious game reduce stress supporting parent application. The New...