Leveraging AI & machine learning to optimize today’s 5G radio access network systems and to build the foundation of tomorrow’s 6G wireless systems

Go back to programme

Leveraging AI & machine learning to optimize today’s 5G radio access network systems and to build the foundation of tomorrow’s 6G wireless systems

  • Watch on zoom

    watch on YouTube

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

  • 46484

    Machine Learning and data-driven models for signal processing have been an extremely exciting space over the past few years and hold enormous promise for state-of-the-art performance in communications systems in complex multi-user and multi-antenna environments and over difficult mediums.

    Wireless applications have long struggled with the tradeoff between tractable optimization problems and model deficit, where our models for the propagation environment, hardware and antenna effects are not accurate enough to represent the stochastic nature of the environment to optimize to its full potential. Data-driven methods have up-ended many years’ worth of thinking in this domain, and allowed us to accomplish both, leveraging detailed and accurate data and distribution information in deployed wireless systems while also maintaining simple high-level problem definitions and efficient implementations.

    This talk will highlight how AI and ML techniques are embracing data in order to better shape efficient radio access networks today and even more so in future cellular technology. We’ll focus on how the physical layer can be thought of as a data-driven machine learning problem, and what this means for better radio and gNB performance in today’s systems. We’ll also discuss how this could more deeply impact future, beyond 5G radio waveform and protocol design to further increase efficiency and density. We’ll provide an overview of work we’ve been doing at DeepSig and at Virginia Tech in order to help realize these visions and discuss other areas where AI/ML techniques hold enormous promise in helping to optimize future wireless systems. Overall, we’ll highlight how data, data driven competitions such as the ITU AI/ML 5G challenges, and data-driven optimization and validation of approaches are an important avenue for communications engineers to embrace at many levels of the stack as communications systems evolve.

    Speaker(s)
    • i

      Speaker abstract

      Deep Learning in the Physical Layer: Transforming 5G and 6G Performance with Data and AI; Wireless systems, access schemes, and band allocations are becoming increasingly complex and heterogeneous. Deep learning-based solutions within the physical layer allow for a rapid and accurate state of the art approach for both spectrum sensing as well as channel access and physical layer design. In this talk, we’ll provide an overview of the background and enablers for this trend of ML in the physical layer of communications and highlight how we are building software solutions based on these approaches and techniques at DeepSig. OpenRAN based wireless solutions are also becoming increasingly prevalent within 5G RAN deployments and provide a key opportunity for rapid deployment of these technologies and to improve their performance in real-world deployments. We’ll highlight how we are using both DL driven PHY techniques and ORAN software to bring these capabilities into ORAN deployments in order to improve energy efficiency and to improve spectral efficiency, capacity, and user experience. Finally, we’ll discuss where we believe these technologies are leading future RAN technologies and standards, and provide highlights from our own trials and partnerships, highlighting the maturity of the technology and its likelihood for adoption in next generation wireless deployments.
      Tim O’Shea
      Co-Founder & CTO
      DeepSig Inc

    Presentation - Tim O'Shea

    Share this session

    Are you sure you want to remove this speaker?