News

MARVEL project in media (DNEVNIK)

Our colleagues and researchers Nikola Simić and Prof. Dragana Bajovic (MARVEL project technical manager) are featured in one of the major daily newspapers in Vojvodina and Serbia, Dnevnik, where they describe their achievements in EU H2020 MARVEL Project, in particular, their work on multi-modal distributed machine learning using audio-visual data from drones, cameras and microphones in emerging Smart Cities. Congrats for successful completion of MARVEL(ous) project!

Link to the article: https://www.dnevnik.rs/drustvo/dronovi-brinu-o-bezbednosti-u-saobracaju-i-na-ulicama-novosadski-ftn-deo-je-velikog

IEEE ISCAS 2023 – CNN-MAC TC Members

IEEE ISCAS 2023 conference, a flagship conference of IEEE CASS – Circuits and Systems Society has just finished with excellent news for ICONIC centre. Our two colleagues, Natasa Samardzic and Stanisa Dautovic are elected as voting members of IEEE CASS Cellular Nanoscale Network and Memristor Array Computing (CNN-MAC) Technical Committee. Congratulations!

new project “REWARDING” funded by serbian science fund

For the next (at least) two years, we will be working on a Green project “WAter quality monitoRing anD INtelliGence” (REWARDING) funded by Fond za nauku Republike Srbije – Science Fund of the Republic of Serbia. We are looking forward to this joint multi-disciplinary collaboration between two teams from Faculty of Technical Sciences, University of Novi Sad and a team from The Institute for Artificial Intelligence of Serbia in the domain of environmental engineering, water quality modeling and monitoring, sensing, deep rural IoT, UAV-based communications, remote sensing, AI modeling, and related fields.

Photo from the event in the Science Fund of the Republic of Serbia:

Horizon Europe “REMARKABLE” marie curie Staff exchange project started

ICONIC researchers participate in a new European Commission Horizon Europe project “REMARKABLE” (Rural Environmental Monitoring via ultra wide-ARea networKs And distriButed federated Learning) that started from January 2023 and will last for 4 years. The project is a Marie Curie staff exchange project. More information about the project is available here: https://cordis.europa.eu/project/id/101086387.

As the first staff exchange in the project, our PhD student Srdjan Sobot spent two weeks at Beyond Vision in Lisbon as part of REMARKABLE that deals with connecting deep rural communities using UAVs. He also used this visit to present his research on deep rural connectivity using UAVs on Cloud and Internet of Things 2023 conference in Lisbon, the location of host Beyond Vision company.

𝐈𝐄𝐄𝐄 – 𝐒𝐏𝐒/𝐄𝐔𝐑𝐀𝐒𝐈𝐏 𝐒𝐮𝐦𝐦𝐞𝐫 𝐒𝐜𝐡𝐨𝐨𝐥 𝐃𝐚𝐭𝐚 𝐚𝐧𝐝 𝐆𝐫𝐚𝐩𝐡 𝐃𝐫𝐢𝐯𝐞𝐧 𝐋𝐞𝐚𝐫𝐧𝐢𝐧𝐠 𝐟𝐨𝐫 𝐂𝐨𝐦𝐦𝐮𝐧𝐢𝐜𝐚𝐭𝐢𝐨𝐧𝐬 𝐚𝐧𝐝 𝐒𝐢𝐠𝐧𝐚𝐥 𝐏𝐫𝐨𝐜𝐞𝐬𝐬𝐢𝐧𝐠

PhD students Tijana Devaja, Ognjen Kundačina and Vukan Ninković present their research at 𝐈𝐄𝐄𝐄 – 𝐒𝐏𝐒/𝐄𝐔𝐑𝐀𝐒𝐈𝐏 𝐒𝐮𝐦𝐦𝐞𝐫 𝐒𝐜𝐡𝐨𝐨𝐥 2022 𝐨𝐧 “𝐃𝐚𝐭𝐚 𝐚𝐧𝐝 𝐆𝐫𝐚𝐩𝐡 𝐃𝐫𝐢𝐯𝐞𝐧 𝐋𝐞𝐚𝐫𝐧𝐢𝐧𝐠 𝐟𝐨𝐫 𝐂𝐨𝐦𝐦𝐮𝐧𝐢𝐜𝐚𝐭𝐢𝐨𝐧𝐬 𝐚𝐧𝐝 𝐒𝐢𝐠𝐧𝐚𝐥 𝐏𝐫𝐨𝐜𝐞𝐬𝐬𝐢𝐧𝐠”, Banja Luka, Sept- 5-9, 2022. The event brought more than 60 students and 10 lecturers from all over the Europe and USA. Congrats to Gordana Gardašević and her team for excellent organisation. Prof. Dejan Vukobratovic presented a lecture on “Gaussian Belief Propagation: Theory and Applications”, which is available on this link.

New director and leadership of ICONIC

Date: 01st October 2021.

A new leadership of ICONIC centre for the following three years (2021-2024) will be:

Director: Prof. Rastislav Struharik

Communications group leader: Prof. Dejan Vukobratovic
Information Processing group leader: Assoc. Prof. Dragana Bajovic
System Design group leader: Assoc. Prof. Ivan Mezei

Congratulations to newly elected researchers with best wishes for successful evolution of ICONIC centre in the following years.

Paper Published in IEEE CLET

Date: September 2021

Research paper entitled “Autoencoder-Based Unequal Error Protection Codes” with ICONIC researchers Vukan Ninkovic and Dejan Vukobratovic as a coauthors is published in IEEE Communication Letters. This paper is a result of collaboration between ICONIC and Chalmers University of Technology researchers as part of H2020 INCOMING project. Congratulations!

Link: https://ieeexplore.ieee.org/document/9525108

ICONIC and COPELABS team won IEEE VTS UAV Innovation Challenge

Date: 28th June 2021.

After winning the first phase of the competition, we are proud to announce that our PhD students won the second and final stage of UAV Innovation Challenge organized by IEEE Vehicular Technology Society. Student team of ICONIC FTN-UNS team lead by Srđan Šobot and with participation of Dušan Bortnik and Vladimir Nikić, along with Brena Lima, PhD student from COPELABS – University of Lusofona, Lisbon, worked together to develop the solution. The first phase solution is presented in the following video: https://www.youtube.com/watch?v=QU0vkbEcO1w while the final solution is presented as part of the TV Show Science for Economy on this link: https://www.youtube.com/watch?v=3_Px0hAjpVE. Congratulations!!!

ICONIC team in H2020 MARVEL project

Date: June 2021.

ICONIC team lead by Dr Dragana Bajovic is participating in H2020 MARVEL project (Multimodal Extreme Scale Data Analytics for Smart Cities Environments). The group recently started with initial experiments that use drones for observation of crowd behaviour in the city of Novi Sad. Crowd classification based on drone and ground recordings will be performed in the experimental setup. Data collection will be performed in the form of staged recordings designed so as to replicate the crowd-monitoring events of interest for training.

For more information about these experiments, read the following blog on MARVEL website: https://www.marvel-project.eu/smart-events-observation-of-crowd-behavior-in-the-city-of-novi-sad/

Prof Dragana Bajovic lecture at Data Week 2021

Date: 25-27. May 2021.

Data Week is the spring gathering of the European Big Data Value and Industrial AI research and innovation community. As part of the workshop entitled ‘Building the data-agile economy from the demand side: evolving business drivers and digital transformation scenarios’, Prof. Dragana Bajovic presented (MARVEL Scientific & Technical Manager) was invited to present H2020 MARVEL plans to take full advantage of the DataBench Toolbox to define the business and technical benchmarks for the evaluation of the MARVEL solution.

For more information about H2020 MARVEL, please see: https://www.marvel-project.eu/.

H2020 C4IIoT results in IEEE Access

Date: April 2021

Research paper entitled “Deep Learning Anomaly Detection for Cellular IoT With Applications in Smart Logistics” with the team of ICONIC researchers including Milan Lukic, Ivan Mezei, Dragana Bajovic and Dejan Vukobratovic as coauthors is published in IEEE Access. This paper presents a part of results obtained during the team activities in H2020 C4IIoT project. Congratulations!

Link: https://ieeexplore.ieee.org/document/9402912