> ## Content Index > Fetch the complete content index at: https://www.adriadefense.com/llms.txt > Use this file to discover other available public pages before exploring further. # US Researchers Test Passive Magnetic Drone Detection With 98.52% Accuracy - URL: https://www.adriadefense.com/us-researchers-test-passive-magnetic-drone-detection-with-98-52-accuracy/ - Published: 2026-10-25T23:08:30.000Z - Updated: 2026-10-25T23:08:30.000Z - Description: US researchers have tested an AI-powered sensor that identifies small drones through magnetic emissions from their motors, offering a potential new approach to passive counter-UAS detection. - Author: Daniel Opoku - Tags: Research, Drones, Counter-UAS, Defense Industry, Industry, Drone Detection, Florida International University, 22Miami, Ukraine, FPV drones, Silynxcom, United States **Florida International University researchers have demonstrated a low-cost method for detecting small drones through magnetic signals generated by their motors, potentially opening a new sensing option for European counter-UAS manufacturers.** Researchers at Florida International University (FIU) in Miami have demonstrated a passive drone detection technology that identifies unmanned aircraft by analyzing magnetic emissions generated by their electric motors. The experimental system achieved detection accuracy of up to 98.52% in outdoor testing conducted at distances reaching 15 meters, according to findings highlighted in an October 23 university announcement. The development was also [reported by Miami-based business publication 22Miami](https://www.22miami.com/article/fiu-passive-magnetic-drone-detection-research?ref=adriadefense.com), which examined the technology's performance, limitations and commercial development status. Unlike conventional radar, the FIU system does not transmit a detection signal. Instead, it uses a specialized antenna to receive extremely low-frequency (ELF) magnetic emissions produced by the brushless electric motors that power small drones. Machine-learning algorithms analyze these emissions to distinguish drones from background signals and identify different aircraft types. The research, published in *IEEE Sensors Journal*, could provide another sensing approach for defense manufacturers developing protection against small unmanned aircraft, particularly in environments where conventional radio-frequency detection faces limitations. ## Why European Drone Manufacturers Should Pay Attention For European counter-UAS developers, one of the most relevant characteristics of FIU's approach is that it does not rely on intercepting a drone's command-and-control transmission. This distinction is particularly significant as Ukrainian battlefield experience drives demand for detecting fiber-optic-guided FPV drones and increasingly autonomous unmanned aircraft. Such platforms can operate without conventional radio command links, limiting the effectiveness of systems that depend exclusively on detecting those transmissions. Magnetic sensing measures emissions produced by the aircraft's motors rather than its communication equipment. In principle, that could make the approach relevant to electrically powered drones regardless of whether they are controlled through conventional radio links, fiber-optic cables or onboard autonomy. However, FIU has not demonstrated that its experimental system can reliably detect fiber-optic-controlled military FPV drones under battlefield conditions. European companies are already exploring different approaches to this challenge. As previously [reported by AdriaDefense, Israeli manufacturer Silynxcom is developing an AI-powered wearable detector](https://www.adriadefense.com/israeli-silynxcom-develops-ai-wearable-to-detect-fiber-optic-drones/) that analyzes acoustic signatures to warn soldiers of approaching drones, including fiber-optic-guided systems. FIU's magnetic sensing research represents a different technological approach that could eventually complement acoustic, optical, radio-frequency and radar detection within integrated counter-UAS architectures. ## Researchers Identify Drone Types Using AI The research project is led by Professor Arjuna Madanayake of FIU's Department of Electrical and Computer Engineering, working with researchers including Chatura Seneviratne and Soumyajit Mandal of Brookhaven National Laboratory. The team's scientific paper, titled *ELF Passive Radio Sensing and AI-Perception of Micro-UAS*, describes a sensing architecture combining a circular loop antenna, low-noise electronics, digital signal processing and a convolutional neural network. Experiments involving three commercial micro-UAS platforms produced detection accuracy of up to 98.52%. The study also reported 83.5% overall accuracy in a four-class identification task involving three drone types and a no-drone category. The distinction matters for defense applications: detecting the presence of an approaching drone and correctly identifying its type are separate technical challenges. The study also found weaker performance for the smallest aircraft, reflecting its lower magnetic emissions. According to FIU, the experimental equipment cost only a few hundred dollars to develop, although this does not establish the manufacturing cost of a commercially deployable product. ## Potential Applications for Counter-UAS Developers The technology's present detection range of approximately 15 meters is its most significant limitation. It cannot currently replace the longer-range surveillance and tracking capabilities provided by military radar and other established counter-UAS sensors. Nevertheless, the research points toward potential applications in close-range protection, including individual defensive positions, perimeter monitoring and localized sensor networks. Further development could explore whether magnetic sensing can be integrated into compact detection devices or combined with existing systems to improve classification confidence. For manufacturers, relevant engineering questions include detection range, performance in electromagnetically noisy environments, classification of unfamiliar drone types, sensor size and the ability to integrate magnetic observations into existing command-and-control software. Such integration is already a major development direction in Europe. Estonian company Defsecintel Solutions, for example, is [working with Estonian-Ukrainian DoD Solution on combining multi-sensor counter-UAS capabilities with AI-based autonomy](https://www.adriadefense.com/defsecintel-dod-solution-link-ai-to-counter-drone-systems/). FIU's work suggests an additional sensor modality that manufacturers could evaluate alongside established detection technologies, rather than a ready-made replacement for them. ## FIU Working With US Industry on Further Development FIU confirmed that its research group is working with US industry to advance the technology, including efforts to improve detection range. The university has not identified a commercial partner, announced a military customer or disclosed a deployment schedule. The published results remain experimental, and broader testing would be required before the technology could be assessed for operational military applications. For European drone and counter-drone manufacturers, the research provides an early indication of how passive motor-signature sensing could become another element of future layered drone defense systems. --- **Sources:** [22Miami: FIU Researchers Report Passive Magnetic Method for Detecting Drones](https://www.22miami.com/article/fiu-passive-magnetic-drone-detection-research?ref=adriadefense.com) | [Florida International University research announcement](https://news.fiu.edu/2026/fiu-engineers-develop-stealthy-low-cost-tech-to-detect-drones?ref=adriadefense.com) | [IEEE Sensors Journal research paper, DOI: 10.1109/JSEN.2026.3665766](https://doi.org/10.1109/JSEN.2026.3665766?ref=adriadefense.com)