RF Threat Warning & Analytics

Cyber Threat and Intrusion Detection with AI-Based RF Sensing


Using DeepSig’s state of the art RF sensing products, OmniSIG Sensor and OmniSIG SDK, organizations are able to perform large-scale RF signature detection and analysis to gain increased awareness of the surrounding wireless environment in a way that was previously prohibitive. These tools enable fast and reliable identification of harmful interference, increased situational awareness of wireless signal anomalies, and analysis of high level wireless behaviors which would be difficult to obtain in the past. Using the OmniSIG SDK, this capability can be custom-tuned for a wide range of custom RF signatures and emitter behaviors. In addition to purely supervised RF emitter detection OmniSIG also provides a powerful set of tools to recognize new and anomalous RF emissions which have not previously been seen. For example below we illustrate “unknown emitter” detection on a 5G-NR signal using an inference model never specifically trained for that emitter type.5g_unknown


By generating large volumes of structured emission detection data at high throughputs, OmniSIG Sensor makes it possible to rapidly detect emitters across a wide range of bands and emitter types while on small or mobile platforms or while deployed on radio infrastructure devices making it an ideal enabler for coverage mapping, usage mapping, interference hunting, unauthorized emitter hunting, cyber-threat detection, and other mobile mapping applications. Due to OmniSIG’s inference times, typically in the low single digit milliseconds, mobile platforms can conduct mapping at unprecedented speed with minimal down time, expense or risk and can detect anomalous or threatening behavior which only occurs for brief fleeting moments in the spectrum or occurs as long reaching trends over time.



By making use of the OmniSIG Sensor’s metadata output in real-time, customers can rapidly realize near-immediate reporting of threats and anomalies and can analyze long term trends and patterns at both fine and long time-resolution over millions of RF events in a manageable way. By turning complex gigabit or more sized firehoses of raw RF sensor data containing numerous wireless spectrum events and trends into compact structured data streams requiring bandwidth orders of magnitude less, data, anomaly detection, pattern recognition and threat detection can be conducted in network and computationally efficient ways. This edge sense-making capability turns RF trend analysis into a structured data problem where tools that are widely used for network and other structured trend analysis can rapidly be leveraged on the domain for tracking, alerting and reporting for wide ranges of spectrum activities.


DeepSig is building specialized web-based analytics tools, and dashboard capabilities leveraging industry standard open source tools to analyze and create understanding of the spectrum through analysis of OmniSIG’s sensing output.

The speed and accuracy of the detection and classification provided by OmniSIG enabled immediate action to new threats, and can be used as input to other automatic cyber defense or alert systems and it can be used to provide analysis, visibility and change detection in cellular and enterprise wireless networks across wide range of wireless technologies including LP-WaN, Cellular, non-standards based or other wireless technologies. Unlike many existing solutions in this space which are closely locked into specific bands or modem technologies, OmniSIG provides a virtually unlimited and comprehensive, actionable new view into the radio spectrum.

We are pioneering the next phase of wireless. 

Interested in learning more about DeepSig and our solutions? Contact us today!