und DeepSig Inc.
„AI-Native Wireless: Leveraging More Data in Physical Layer Design“
In recent years, taking a new data-driven approach to designing parts of the wireless physical layer alongside existing model-driven approaches has provided exciting opportunities for further optimization using more information than previously possible. Thanks to advancements in training and inference algorithms, neural architectures, and inference silicon, it's now more feasible to deploy these algorithms with low latency and power consumption, making them increasingly competitive for wireless signal processing.
Our team at DeepSig has been focused on leveraging both data-driven and model-driven design to rapidly understand the spectral and spatial world at the edge and to optimize baseband signal processing functions to improve power and cost efficiency within current standard implementations. There's also been broad discussion of future AI-Native Air Interfaces in 6G wireless which could involve tasks such as data-driven adaptive modulation, pilot, frame and even source and channel coding design as well as limited AI-Native standard functions such as CSI feedback encoding in the 5G Advanced.
In this talk, we'll discuss the enablers of this trend, showcase key results and capabilities attained so far, highlight proof of concept capabilities and technologies which may be further adopted in the future, and explore what to potentially expect in the next 5-10 years in the evolution of AI-Native communication systems and their potential impact.
is the CTO and Co-Founder at DeepSig Inc and a Research Assistant Professor at Virginia Tech in Arlington, VA. Tim O’Shea
He is focused on leveraging machine learning and data-driven approaches within the wireless physical layer to help improve baseband processing spectral efficiency, energy efficiency, and environmental awareness and automation.
Previously he worked with wireless startups Hawkeye 360 and Federated Wireless in seed stage and held engineering R&D positions with both the US DOD and with Cisco Systems. He is the author of over 100 academic works and patents in this area and is involved in IEEE COMSOC, IEEE MLC ETI, Next-G Alliance, and OpenRAN efforts to accelerate AI driven communications system technology and their adoption within next generation RAN.