IDLab

New smart mobility project SARWS focuses on dangerous weather conditions

2019-02-04T00:00:00+00:00

A consortium of research and business partners, led by IDLab and Verhaert, launched a new project to improve road safety, called Secure and Accurate Road Weather Services (SARWS). In this project sensor data of cars will be used to generate real-time warnings about local weather conditions, like hail or poor visibility, and dangerous road conditions, like ice or snow. The project is financed by VLAIO (Flemish government) and the participating companies. You can read more about SARWS on the KMI website (dutch).

New supercomputer boasts unparalleled computing power

2019-01-30T00:00:00+00:00

It may not look spectacular, but its performance certainly is: with their acquisition of an NVIDIA DGX-2™, the University of Antwerp and imec have become the first university in the Benelux to own the latest generation of supercomputer. “This computer has a calculation speed of two petaflops per second,” explains Steven Latré. “One petaflop is 10 to the 15th power, or a quadrillion.”

You can read more about our new supercomputer on the IDLab UA website.

IDLab team wins a 2nd award of 750,000 USD in Phase 2 of the DARPA Spectrum Collaboration Challenge

2018-12-13T00:00:00+00:00

For the second time in a row, team SCATTER with researchers from imec-IDLab and Rutgers University (US), have won a prize in the DARPA Spectrum Collaboration Challenge, the world’s first collaborative machine-intelligence competition to overcome spectrum scarcity. Team SCATTER, led by Ingrid Moerman (Ghent University), Steven Latré (University of Antwerp) and Ivan Seskar (Rutgers University) is one of the six prize-winners of the second phase of the competition. Again team SCATTER team has shown that its wireless network solution with intelligent, autonomous radios is capable to collaborate with competing teams in challenging dynamic environments with high traffic demands. Team SCATTER will now continue in the third and last phase of the competition which will end with a live finale at Mobile World Congress Americas Los Angeles in Oct 2019.

More information can be found on the DARPA’s Spectrum Collaboration Challenge website:

IDLab deploys testbed for connected cars

2018-12-11T00:00:00+00:00

IDLab will deploy a testbed for communication between connected cars on the E313 highway in Antwerp, as part of the Flemish Smart Highway and the European CONCORDA project. This technology will also assist the self-driving car of the future.

Bart Lannoo from IDLab Antwerp recently explained on the Flemish television what IDLab is doing in the Smart Highway and CONCORDA project.

IDLab team wins award in first Preliminary Round of DARPA Spectrum Collaboration Challenge

2018-01-12T00:00:00+00:00

The team from IDLab, lead by Prof. Dr. Ingrid Moerman and Prof Dr. Steven Latré, won $750,000 in prize money on the first event of the DARPA’s Spectrum Collaboration Challenge (SC2), which is the world’s first collaborative machine-intelligence competition to overcome spectrum scarcity. The SC2 is a three-year long tournament designed to generate new wireless paradigms and access strategies in which radio networks enhanced with artificial intelligence (AI) will autonomously collaborate and reason about how to share the increasingly congested electromagnetic (EM) spectrum.

Read more about SC2 in the DARPA press release.

New spin-offs

2017-11-09T00:00:00+00:00

Two new start-ups were founded with the support of IDLab.

BiFast

Bifast develops high speed I/O chips. They offer both ASIC and standard products. Their flagship product is a 110Gbps transceiver chipset for data center communications over twin axial copper cables.

This project has been funded by the Industrial Research Fund at Ghent University and was established with the support of the Research Group on Electromagnetism, RF and High-Speed Circuits and Systems within UGent’s ID-Lab.

ML2Grow

ML2Grow offers artificial intelligence based predictive tools that enhance decision making when optimizing processes and systems, starting from the customers data and model. One example is for instance yield prediction tools for greenhouses. This spin-off was established with the support of Machine Learning and Data Mining Research Group of UGent’s IDLab.

H2020 WiSHFUL Open Call for Experiments

2017-04-04T00:00:00+00:00

The WiSHFUL project, coordinated by Ingrid Moerman from imec/UGent/IDLab, hereby announces its fourth Open Call for Experiments, targeting advanced solutions for controlling wireless networks using the WiSHFUL software platform and unified programming interfaces (UPIs), and using the facilities and hardware supported by the WiSHFUL Consortium. WiSHFUL has extended several wireless radio platforms with control plane functionality, and as such offers a unique experimentation environment for early implementation and validation of end-to-end 5G solutions that improve resource utilization through advanced reconfigurabilty of radio and network settings. WiSHFUL further provides all the experimentation tools to validate 5G solutions and can adopt new 5G technologies from external experimenters and support new technologies within a minimum amount of time. Successful proposals will receive funding, will have guaranteed access to the testbeds and obtain elaborated support by the WiSHFUL consortium.

Detailed call information (call document & proposal template) can be found at the WiSHFUL project website.

imec contributes to Open Framework for Multilingual and Semantic Enrichment of Digital Content

2017-03-08T00:00:00+00:00

After two years, FREME has just come to the end of its fruitful development road. The project – co-funded by the European Commission under the H2020 Framework Programme for Research and Innovation – has developed a commercial-grade open framework of e-Service for the multilingual and semantic enrichment of digital content. This framework was designed and implemented thanks to the effort of some specialists in language resources, tools, and services, multilingualism and Semantic Web, Linked and open data, belonging to paramount actors such as DFKI, Infai, and imec.

In the first year, imec was responsible for the development of the e-publishing service that an API to create digital books. In the second year, imec acquired both feedback and a high number of potential customers for the FREME framework. Furthermore, we developed a user interface that allows easy access to the different e-services for new and experienced users. This has resulted in a very positive final review. imec has received “excellent” scores in all areas, with explicit positive mentions of its role as Technology Provider in the second year.

Starting from February 2017, FREME will be hosted for at least two years by the Irish ADAPT Center, a dynamic research centre that combines the world-class expertise of researchers at four universities (Trinity College Dublin, Dublin City University, University College Dublin and Dublin Institute of Technology) with that of its industry partners to produce ground-breaking digital content innovations. The aim of the synergy between FREME and ADAPT is to seek to further develop the framework with the goal of empowering both research and industry projects.

IDLab researcher co-founds Indigo Diabetes

2017-01-23T00:00:00+00:00

Indigo Diabetes, a spin-off from Ghent University and imec, develops a silicon-photonics needle-free glucose sensor to measure and manage blood glucose levels without the need for blood droplets or finger pricks. The basic concepts and key IP emerged from the Photonics Research Group, where Danaë Delbeke incubated the technology in collaboration with two other research groups in Ghent, CMST and IDLab. While CMST focused on the development of packaging solutions, IDLab focused on the very sensitive, but low-power read out electronics and signal processing.

IDLab researchers winners in worldwide data science competition

2016-05-02T00:00:00+00:00

Transforming How We Diagnose Hearth Disease

At IDLab, we are leveraging our data science expertise to transform how we diagnose heart diseases.

Our researchers finished among the winners of US 2nd National Data Science Bowl (on Kaggle.com) with a machine learning algorithm to automatically determine cardiac volumes from MRI scans. In 2015, this team also won the “Deep Sea” challenge , out of 1049 teams, in the 1st National Data Science Bowl.

The goal was to estimate minimum (end-systolic) and maximum (end-diastolic) volumes of the left ventricle from a set of MRI-images taken over one heartbeat, as these volumes are used by practitioners to compute an ejection fraction: fraction of outbound blood pumped from the heart with each heartbeat. This measurement can predict a wide range of cardiac problems. For a skilled cardiologist analysis of MRI scans can take up to 20 minutes. Automating this process enables MRI scanners to evolve from imaging devices towards cloud-assisted, smart tools in the process of diagnosing heart diseases.