Presenting TRALICO at the EIG CONCERT-Japan Kick Off Workshop

Our departmental colleagues Vilmos Simon, Árpád Huszák and Bereczki Norman, as well as Ismail Arai, a researcher from our Japanese partner, the NAIST, presented the TRALICO (Multi-Input Deep Learning for Congestion Prediction and Traffic Light Control) project at the EIG CONCERT-Japan Kick Off Workshop in Prague, with presentation and poster.

Our department’s MEDIANETS lab, together with partners Nara Institute of Science and Technology (Japan) and Istanbul IT and Smart City Technologies Inc. (Turkey), are developing solutions that can predict future traffic intensity and emerging traffic congestion using artificial intelligence. These predictions will be used to intervene in the control of traffic lights in real time, adjusting them to the predicted traffic characteristics, reducing travel time and emissions.

Presenting to the chair
Vilmos Simon presenting

Our colleagues’ participation at the TRALICO project meeting

Our departmental colleagues Vilmos Simon, Árpád Huszák and Norman Bereczki participated in the meeting of the TRALICO (Multi-Input Deep Learning for Congestion Prediction and Traffic Light Control) project in Istanbul, holding discussions with their consortium partners, the Nara Institute of Science and Technology (Japan) and the ISBAK – Istanbul IT and Smart City Technologies Inc. (Turkey).

They visited the research and development center of ISBAK and the traffic control center of Istanbul Metropolitan Municipality, where the traffic of 16 million people is monitored and managed every day. Together with their partners, they develop solutions based on artificial intelligence to predict future traffic intensity and emerging traffic congestions. These predictions will be used to intervene by traffic lights control in real time, adjusting the programs to the predicted traffic characteristics, reducing travel time and emissions.

More photos from the event:

TRALICO meeting
TRALICO visit
TRALICO center
TRALICO setup

New PhD students in the lab

We proudly announce that this September three new PhD students have joined the MEDIANETS team.

Yasameen Sajid earned her MSc degree from Al-Qadisiyah University, Iraq, in 2021. From 2021 to 2024, she served as an Assistant Lecturer. In 2024, she was awarded the prestigious Stipendium Hungaricum Scholarship, enabling her to pursue PhD studies at the Budapest University of Technology and Economics (BME). Her research focuses on network routing, energy efficiency in wireless sensor networks (WSN), and network traffic and event classification.

Hassanien graduated from the University of Technology and Engineering as a Computer Engineer in Baghdad (Iraq) in 2006, then he got a master degree from Reading University (Advanced in computer Science) in the UK in 2014. In September 2024, he got Stipendium Hungaricum Scholarship to join the PhD program at BME MEDIANETS Lab. His primary research interests are machine learning, reinforcement learning, traffic management for autonomous vehicles.

Mohammed Mohaisen received his MSc in Business Informatics Engineering from Corvinus University of Budapest in 2023 and holds a BSc in Computer System Engineering from Al-Azhar University. He is currently a PhD student at the BME MEDIANETS Lab under the Stipendium Hungaricum Scholarship, where he also serves as a laboratory assistant for the Basic Programming 1 course. His research focuses on machine learning-based traffic optimization in smart cities and V2X communication.

Our colleague László Bokor participated in the annual Car-2-Car forum

As a result of the invitation of our industrial partners, our colleague László Bokor participated in the Car2-Car Communications Consortium’s 2024 Forum in Frankfurt. The 2024 meeting of the professional and scientific community of automotive communications on October 22-23, 2024, in Frankfurt, Germany, highlighted several hot topics, such as actual deployment questions of cooperative intelligent transport systems, relevant activities of the automotive industry, the current state of the art in the field of protection of vulnerable road users and optimal usage of radio resources. László Bokor was involved in discussions of C2C-CC activities requiring advanced simulation modeling of V2X communication technologies and solutions.

Our Head of Lab, Vilmos Simon was elected by the HUN-REN Headquarters Governing Board as a member of the Scientific Council, among the members of the College of Mathematics and Natural Sciences

HUN-REN was established to manage and operate a centrally funded, independent research network that is a cornerstone of Hungarian scientific life. The researchers of its eleven research centers and seven research institutes, as well as its 116 supported research groups operating at universities and other public institutions, are in the fields of mathematics and natural sciences, life sciences, and they carry out basic and applied research in the most diverse areas of the humanities and social sciences. The Scientific Council supports the work of the Governing Board, comments on proposals of strategic importance prepared for the Governing Board, prepares strategic proposals and carries out an annual evaluation and decides on the powers delegated to it by the Governing Board.

“Best Student Paper Award” received by Péter Pálos

Our student, Péter Pálos received Best Student Paper Award from the Committee of 44th International Conference on Telecommunications and Signal Processing (TSP). The honored paper titled “ReLight-WCTM: Multi-Agent Reinforcement Learning Approach for Traffic Light Control within a Realistic Traffic Simulation” more…

Presentation at VAE 2018

Our colleague Zsolt Szendrei presented the article “Zs. Szendrei , N. Varga, L. Bokor: A SUMO-based Hardware-in-the-Loop V2X Simulation Framework for Testing and Rapid Prototyping of Cooperative Vehicular Applications” on the 23rd of May 2018 at the 2nd International Conference on Vehicle and Automotive Engineering – 23-25 May 2018, University of Miskolc, Hungary (VAE 2018).

  

Abstract. Vehicle-to-Anything (V2X) technologies aim at providing globally standardized communication tools to efficiently transmit information between all players of the transportation ecosystem. Based on such extensive data exchange between traffic objects V2X helps to create an advanced domain of transportation called Cooperative Intelligent Transportation Systems (C-ITS) where an ever-growing scale of cooperative vehicular applications/services facilitate enhanced safety and comfort on the road and lead towards fully automated transportation in the future. C-ITS relies on a complex architecture consisting of a cross-layer op-timized sophisticated protocol stack, hybrid radio access solutions, computing al-gorithms, decision schemes, special interfaces, internet-of-things (IoT) integra-tion, etc. Moreover, C-ITS and the support of cooperative use-cases demand high reliability, enhanced Quality of Service/Quality of Experience (QoS/QoE), and rock-solid hardware/software implementations working efficiently and securely even in the most complicated environments including extreme traffic circumstanc-es or unpredictable actions. Therefore, deliberate system testing is a crucial and strategic process, especially when we are closing to the wide-scale deployment phase of real-life C-ITS solutions. Our proposed hardware-in-the-loop (HiL) V2X simulation framework was designed to offer a cost-efficient and simple toolset for testing and rapid prototyping of cooperative vehicular solutions by partially replacing costly, time-consuming and oftentimes dangerous field tests with an easy to install, tabletop laboratory test- and development suite.

Presentation in Žabljak – IT 2018

Our colleagues Norbert Varga and László Bokor together with co-authors Ivana Ognjanović and Ramo Šendelj from Montenegro presented their accepted paper entitled “On the mobility management performance of NEMO-based nested wireless body area networks” in Žabljak at the 23. Information Tehnology (IT 2018) conference.

Abstract:
Given the currently fast evolution of wireless access technologies, high-performance mobile devices and wide variety of wearable medical sensors, electronic and mobile healthcare (eHealth and mHealth) services are receiving particular attention nowadays. Several mHealth application scenarios strongly rely on mobile or embedded devices interconnected with medical sensors via Wireless Body Area Networks (WBANs). The vast majority of these services require strict medical level Quality of Service (QoS) and Quality of Experience (QoE) provision. Emergency use-cases like remote patient monitoring, tele-consultation and guided surgical intervention demand real-time communication and appropriate connection quality during any mobility event. In many cases the mHealth WBANs are nested into other overlay access networks around the patient. To fulfil the quality requirements of such applications, during the various mobility scenarios not only a single mobile entity has to be considered, but also entire nested WBANs (moving between different networks) have to be taken into account. If network layer mobility management is required (like in case of inter-technology handovers under heterogeneous radio coverages), still NEMO Basic Support (RFC 3963) is the most common standardized approach to be implemented. Motivated by this, the paper investigates mobility management performance issues of NEMO-based nested WBANs. The evaluation is based on INET/OMNeT++ simulations and transport layer characteristics.