DeepSig Blog

Reinventing Wireless with Deep Learning

DeepSig Talks Wireless AI at NVIDIA GTC

Posted by Tim O'Shea on Apr 1, 2021 10:20:28 AM

DeepSig’s CTO Dr. Tim O’Shea will be presenting during two sessions focused on wireless AI/ML at NVIDIA’s GTC 21 conference taking place online April 12-16, 2021. NVIDIA's GTC conference brings together a range of AI innovators, technologists, and creatives looking to ignite ideas and forge new connections to take on our biggest challenges.  We're excited to share a range of key updates on how we are using AI/ML at the edge to accelerate wireless sensing, 5G vRAN, and drive the future of communication systems!

We will cover a number of exciting new updates and results from our work in accelerating neural network inference on baseband radio signals for use in wireless signal sensing, as well as 5G edge networks, and beyond-5G wireless networks.   Among these updates, we'll share a number of key results on how our usage of AutoML, pruning, fusion, reduced precision inference, and other techniques can provide massive improvements to these applications in terms of model speed and accuracy.  We'll be sharing a number of exciting updates to our OmniSIG and OmniPHY 5G software capabilities including showing inference scaling to hundreds of MHz and GHz-wide bandwidth processing capabilities and a range of additional new signal types and features.   

DeepSig is pioneering the use of deep learning to redefine the state of the art in the performance and efficiency of signal processing and radio systems.  By developing new data-driven methodologies, tools, and software for the design and optimization of wireless communications algorithms which better match real world effects using efficient neural network inference, DeepSig can achieve unparalleled results in system performance, cost, and power efficiency.  DeepSig provides highly optimized software components for use in radio systems, virtualized RAN, and other platforms, to accelerate the growth and adoption of intelligent wireless system functions at scale in production.

Click the link to register free for NVIDIA GTC and join us for Dr. O’Shea’s presentations, plus gain access to 1500+ live and recorded sessions, interactive panels, demos, research posters, and more. Learn more at

DeepSig Inc.
is a venture-backed and product-centric technology company developing revolutionary wireless processing software solutions using cutting edge machine learning techniques to transform baseband processing, wireless sensing, and other key wireless applications.

New call-to-action