Digital-twin-enabled 6G: Depth Map Estimation in 6G mmWave systems

Go back to programme

Digital-twin-enabled 6G: Depth Map Estimation in 6G mmWave systems

  • Watch

    * Register (or log in) to the AI4G Neural Network to add this session to your agenda or watch the replay

    • Watch

      Register

      * Register (or log in) to the AI4G Neural Network to add this session to your agenda or watch the replay

    In future 6G networks, digital twins will virtually implement the physical wireless propagation environment, enabling learning, optimization, and dynamical re-calibration of 6G operational parameters to improve network performance. To fulfil this vision, extracting new information, such as depth maps of an environment, from existing sensors is of greatest importance to enable and create scalable and efficient digital twin networks. Using existing mmWave systems already integrated to nowadays devices incurs no additional cost compared to adding new sensors with extra-capabilities. Jointly using communication signals to perform depth map estimation, enables easier network management, keeping network bandwidth usage, reliability, and latency under control, since no extra data and overhead is generated by using secondary sensors. Equally importantly, exploiting signals already designed with the purpose of wireless communication will avoid energy consumption escalation.  

      

    This talk introduces the problem statement “Depth Map Estimation in 6G mmWave systems” for the 2023 ITU AI/ML in 5G Challenge. This is the second edition of the challenge; after the success of the first edition, we challenge participants to apply ML techniques to outperform the baseline solution provided. Learn how the NIST Communications Technology Laboratory is leveraging innovative measurement methods and equipment to shed light on millimeter wave propagation in real world environments and how Samsung is developing ML models to tackle the challenges of future wireless systems.

    Share this session
    • Watch

      Register

      * Register (or log in) to the AI4G Neural Network to add this session to your agenda or watch the replay

    • Start date
      15 May 2023 at 14:00 CEST Geneva | 08:00-09:00 EDT, New York | 20:00-21:00 CST, Beijing
    • End date
      15 May 2023 at 15:00 CEST Geneva | 08:00-09:00 EDT, New York | 20:00-21:00 CST, Beijing
    • Duration
      60 minutes
    • Programme stream
    • Topics
    • UN SDGs