Converge Challenge: Multimodal Learning for 6g Wireless Communications
Abstract
High-frequency mmWave and sub-THz systems enable ultra-high data rates but suffer from severe path loss and blockage sensitivity. Visual sensing can enhance reliability by providing environmental awareness for proactive beam management, yet progress has been limited by the lack of synchronized real-world multimodal datasets. This CONVERGE challenge addresses this gap with a novel indoor mmWave dataset and three tasks: Blockage Prediction, UE Localization, and Channel Prediction. These tasks are designed to benchmark cross-modal learning and promote collaboration between the wireless and computer vision communities.
Fields of Study & Subjects
Speech and dialogue systems
ICT in Developing Communities
Speech Recognition and Synthesis
Publication details
- Journal / source
- ICASSP 2026 - 2026 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
- Publisher
- IEEE
- Pages
- 21799–21801
- Publication date
- 21 April 2026
- Language
- EN