The Rise of AI for Defense Applications
An army marches on its stomach. Napoleon didn’t actually say that, but the point remains: A well-fed army is better able to win than one that’s malnourished.
Today, defense organizations increasingly rely on artificial intelligence for analysis and action both on and off the battlefield. To win, AI needs a steady diet of information. Antennas play a critical role by ensuring it has a reliable, high-performance connection to myriad sources of information.
To enable AI, antennas are increasingly using it, too. One example is GNSS antenna systems that use AI to detect jamming and spoofing, which literally throw fighter jets, military drones, and other aircraft off course.
In spoofing attacks, false signals deceive the aircraft’s GNSS receiver, making it believe that it’s in a different location or at a different point in time. By relying on these false signals, the receiver provides incorrect timing and positioning data, with potentially severe consequences. Jamming attacks overpower the aircraft’s GNSS receiver with strong RF signals in the same or adjacent frequency bands to the legitimate satellite signal.
How AI Mitigates GNSS Jamming and Spoofing
AI enables new defenses by detecting and classifying different types of jamming and spoofing attacks. One example is ATLANTIS, an EU-funded research project that uses AI to detect anomalies in signal strength, frequency spikes, and discrepancies in signal arrival times.
“AI’s learning capabilities allow these systems to continuously evolve and respond to new types of jamming and spoofing attacks,” ATLANTIS researchers explain in their AI-Based GNSS Jamming and Spoofing Detection and Classification paper. “As attackers develop more sophisticated methods, AI-based detection systems can learn from these new patterns, ensuring they remain effective in identifying and mitigating threats. This ongoing learning process is pivotal in maintaining the integrity of GNSS services in an ever-changing security environment.”
Another research project is using AI to combat attacks that use jamming and spoofing simultaneously. “In the GNSS/GPS jamming detection task, we attained approximately 99% accuracy, improving performance by around 5% compared to previous studies,” the researchers wrote in GNSS/GPS Spoofing and Jamming Identification Using Machine Learning and Deep Learning.
AI: Antenna Intelligence
The common denominator with these new defenses is that they require high-quality antenna systems to collect all of the necessary data for the AI to determine whether an attack is underway. These AI-powered defenses also enhance the effectiveness of existing defenses, where antennas also play a critical role.
Take the example of in-band jamming. A Controlled Reception Pattern Antenna (CRPA) system uses multiple antenna elements to null out the interfering signals so the receiver can focus on the legitimate GNSS signals.
Another example is leveraging multiple signals from multiple constellations. This gives the GNSS receiver alternatives when the AI determines that primary GNSS signals are jammed. A multi-constellation defense also can be combined with a CPRA to support all frequency bands simultaneously and perform independent beam nulling in each of those bands. (For a deeper dive, see “Get Out of a Jam: How GNSS Antennas Help Thwart Jamming Attacks” and “Countering GNSS Jamming and Spoofing for Aerospace and Defense Applications.”)
Get in touch for orders or any queries: sales@rfdesign.co.za / +27 21 555 8400
Courtesy of Taoglas
