Early-Stage Company’s Deep Learning Software Reinventing Wireless Communications

Different Wireless Communications

One subset of artificial intelligence (AI) that is disrupting common technological modalities as we speak is machine learning (ML), which is revolutionizing sectors such as online fraud detection, video surveillance, online customer support systems, and others.

Yet, it’s going way beyond that, by even reengineering wireless communications.

The performance goals for 5G are ambitious, impressive, and will be world-changing. Achieving them is also a significant challenge, and as the world races to develop, test, and deploy 5G technology, existing approaches using traditional design methodologies are starting to show some of their limits

 — DeepSig

“DeepSig has unique domain experience in both the government and the commercial sectors, and we’re rapidly evolving our open standards-based AI software to support a growing range of wireless systems and use cases,” said James Shea earlier this year. Shea is the CEO of DeepSig, a startup he co-founded along with Tim O’Shea back in 2016.


A venture-backed company pioneering the application of deep learning to reinvent wireless communications by replacing core wireless technology with it, DeepSig is creating communication systems that are faster, more cost-efficient, more secure, and able to excel in complex environments.

The startup’s critical success factor is due to its revolutionary approach to signal processing. DeepSig employs ML and AI to learn optimized models directly from data rather than manually designing specialized algorithms.

This method is an effective substitute for hand-engineering algorithms with AI-based equivalents that exploit ML, ultimately delivering more computational and power benefits to many areas of the 5G stack.

Based in Arlington, Virginia, DeepSig’s products include the OmniSIG™ Sensor, which provides a new class of RF sensing and awareness; the OmniSIG™ SDK, a tool suite allowing customers to curate RF datasets, train state-of-the-art deep learning inference models for custom wireless sensing applications, and deploy them to edge-sensing devices; the OmniPHY-5G™, enhancing 5G RAN performance with ML, as well as the OmniPHY™.

DeepSig’s CEO and co-founder James Shea is an experienced engineer, entrepreneur and executive with a successful track record of business startups, turnarounds, and the ability to drive innovation in both commercial and military markets. Skilled at assembling the proper management, technical, and financial talent to grow profitable technology companies, his wide breadth of experience spanning commercial wireless, military electronics, and test and measurement markets is a crucial factor in DeepSig’s continual growth and dynamism.

He obtained an MS in Electrical Engineering from the University of Illinois Urbana-Champaign.

Tim O’Shea is the CTO and other half of the founding team. His research on how to apply deep learning to wireless signal processing, part of his Ph.D. work in Electrical Engineering at Virginia Tech, led to breakthroughs in software radio, ML and network security.

With some $7.7 million in funding over three rounds according to Crunchbase data — the latest funding coming from a $5-million Series A round led by investor Leawood Venture Capital in early 2020 — DeepSig has the capital to pioneer deep learning for wireless communications and signal processing, and in turn, make our lives much easier.