Workshop 28
AI native Communications Systems - towards Integrated Intelligent and Highly Efficient Communications Systems (IIHECS)
Scope and topics of the workshop
The rise of Artificial Intelligence (AI) and Machine Learning is a game change on almost every level of industry. The field of wireless communications, and specifically the evolution of cellular mobile communications towards its 6th Generation as currently under definition in the 3rd Generation Partnership Project (3GPP), is no different and multiple strategies are currently being investigated on how to take advantage of this technology. Initial targets include the optimization of Multiple-Input-Multiple-Output (MIMO) processing, advanced localization techniques, improved Channel-State-Information (CSI) feed-back management and others. Besides specific technical optimizations, the arrival of AI/ML questions the basic design approach of wireless communication systems and may require rethinking the overall approach to architecture for exploiting AI/ML principles within the network itself and externally, through so-called AI-as-a-Service (AIaaS) based solutions. Besides the technical challenges, Government Agencies around the world are starting to regulate the usage of AI/ML technology; examples include the US White House Executive Order on the Safe, Secure, and Trustworthy Development and Use of Artificial Intelligence, the European Artificial Intelligence Act, the South Korean Artificial Intelligence Basic Law, etc. It indeed needs to be understood which constrains such regulations put onto the design of Communications Systems and which solutions need to be implemented in order to achieve compliance. Related solutions may be designed for optimized operation in central or distributed processing units or the edge.
The scope of this workshop is to explore this twofold role of the technical exploitation of AI/ML technology in wireless communications systems and the management of the related regulation framework arising around the world. Contributions are sought for on all levels of abstractions, including innovative architectural approaches, the usage of AI/ML principles to further optimize existing processing paradigms and pushing the boundaries of efficiency, solutions supporting regulation requirements of Government administrations and other.
Topics of interest include, but are not limited to:
- Architectural solutions integrating AI/ML in central, distributed or edge processing
- Efficient and effective network architectures for learning and inference
- AI/ML in 6th Generation Cellular Mobile Communications Networks
- AI/ML in Radio Local Area Networks (RLAN)
- AI/ML for cross-layer edge resource management and orchestration
- AI/ML for improving physical layer processing of communication systems
- Distributed intelligence and federated learning for optimization of communications networks
- Modeling and performance analysis for communications learning systems
- Energy efficiency considerations related to AI/ML solutions
- Architectural and other solutions meeting Government Regulation requirements
- Generative AI applied to wireless communications
- Semantic and Goal-oriented communications in the context of AI/ML enabled wireless communications
- Testbeds and performance evaluation of AI/ML-enabled wireless network
Workshop agenda
| Time | Session |
|---|---|
| 08:30–08:35 |
Opening session
|
| 08:35–09:15 |
Keynote 1
|
| 09:15–09:30 |
ML-based Uplink Spatial Compression for O-RAN Split 7.2 in Massive MIMO
|
| 09:30–09:45 |
End-to-end Learning of Probabilistic and Geometric Constellation Shaping with Iterative Receivers
|
| 09:45–10:00 |
Vector Quantization-Based Opportunistic Semantic Communication for Object Detection
|
| 10:00–10:15 |
Compacting Semantic Matryoshka Representations with Product Quantization
|
| 10:15–10:30 |
CNN-BiLSTM Adaptive DDQN for QoS-based NOMA-MEC Offloading in Smart Factory
|
| 10:30–11:00 |
Coffee break
|
| 11:00–11:15 |
Toward Semantic Translation in Network Slicing: LLM-Based Network Slice Configuration
|
| 11:15–12:00 |
Technical Tutorial
Abstract: The European Artificial Intelligence Act (EU AI Act) is a key legislation affecting related debates around the world. We give an introduction on the EU AI Act, summarize prohibited practices and outline technical requirements to be met for so-called “High Risk” AI Systems. We furthermore outline open questions of specific interest to the scientific community, for example related to quantitative assessment of requirements related to the quality of training data, avoidance of biases and other.
|
| 12:00–12:30 |
6GARROW Demo
Short description: The Smart Networks and Services Joint Undertaking (SNS JU) project 6GARROW is an European/Korean research project under the European Union’s Horizon program, Grant Agreement Number 10119219 and by the Institute for Information & Communications Technology Promotion (IITP) grant funded by the Korean government (MSIT) (No. RS-2024-00435652). The Demo will illustrate 6GARROW results related to the integration of AI and Semantic communication capabilities into future 6G Communication systems.
|
| 12:30–14:00 |
Lunch
|
| 14:00–14:45 |
Keynote 2
|
| 14:45–15:00 |
Refining Diffusion-Based OFDM Channel Estimation via Intermediate-State Guided Reverse Process
|
| 15:00–15:15 |
Enhancing Energy Efficiency in Cellular Networks via Power Consumption Model Exchange
|
| 15:15–15:30 |
Semantic Waveforms for AI-Native 6G Networks
|
| 15:30–16:00 |
Coffee break
|
| 16:00–16:15 |
Performance Analysis of Channel Prediction Accuracy with KAN
|
| 16:15–16:30 |
Cooperative Resource Management for Efficient Energy-Traffic Matching in Cell-Free Massive MIMO Systems with Hybrid Energy Supply
|
| 16:30–16:45 |
Semantic Channel Equalization Strategies for Deep Joint Source-Channel Coding
|
| 16:45–17:25 |
Panel discussion
|
| 17:25–17:30 |
Closing the workshop
|
When?
Friday, 12 December 2025
8:30 AM to 5:30 PM
Where?
Taipei, Taiwan
Important dates
Papers submission:
15 July 2025
Acceptance notification:
1 September 2025
Camera ready:
1 October 2025
Workshop:
12 December 2025
Organizing committee
General chairs
- Dr. Emilio Calvanese Strinati, CEA-Leti (France)
- Dr. Markus Dominik Mueck, Intel Deutschland GmbH (Germany)
Technical program chairs
- Prof. Matti Latva-aho, University of Oulu (Finland)
- Dr. Markus Dominik Mueck, Intel (Germany)
- Prof. Seong-Lyun Kim, Yonsei University (Republic of Korea)
Steering committee
- Dr. Vincenzo Sciancalepore, NEC (Germany)
- Dr. Paolo Di Lorenzo, Sapienza University of Rome (Italy)
- Prof. Eryk Dutkiewicz, University of Technology, Sydney (Australia)
Tecnical program committee
- Jinho Choi, Adelaide Univ. (Australia)
- Seung-Woo Ko, Inha Univ. (Korea)
- Jeonghun Park, Yonsei Univ. (Korea)
- Joongheon Kim, Korea Univ. (Korea)
- Jaehoon Chung, LG Electronics, Inc. (Korea)
- Taesang Choi, ETRI (Korea)
- Premanandana Rajatheva, Univ. of Oulu
(Finland) - Nicolas Cassiau, CEA-Leti (Grenoble)
- Alexis Dowhuszko, Aalto Univ. (Finland)
- Riku Jäntti, Aalto Univ. (Finland)
- Zoran Utkovski, Fraunhofer HHI (Germany)
- Pierre Dal Zotto, Grenoble Ecole de Management (France)
- Davide Montagno Bozzone, Hewlett Packard Ent. (Italy)
- Louis-Adrien Dufrène, Orange (France)
- Quentin Lampin, Orange (France)
- Guillaume Larue, Orange (France)
- Matti Latva-aho, Univ. of Oulu (Finland)
- Dileepa Marasinghe, Univ. of Oulu (Finland)