Researchers train AI to detect concealed handguns using Wi-Fi signals
The system uses ambient Wi-Fi signals to identify hidden weapons without requiring direct line of sight. It operates on older 802.11a standards and is available as open-source software.
Etiido Uko, a researcher at the Swedish Defense Research Agency (FOI), has led the development of an AI system that detects concealed handguns using Wi-Fi signals. The system, called WiFireNet, analyzes how objects distort ambient wireless signals to identify the presence of weapons. This technology does not require direct visual access to the object being scanned, making it a potential tool for security applications in various environments.
The research builds on previous work in Wi-Fi-based sensing, where security experts have explored ways to use wireless signals for detecting hidden objects. The system leverages channel state information from Wi-Fi signals, which change when an object is present. By training an AI model on this data, the researchers have achieved high precision in detecting concealed weapons, even on older 802.11a Wi-Fi standards.
The system's performance was tested using the WiFire Dataset, which includes data collected from various scenarios involving concealed weapons. The research team reported that the AI model achieved a high detection accuracy, demonstrating the feasibility of using Wi-Fi signals for security applications. The open-source nature of WiFireNet allows for further development and adaptation by other researchers and organizations.
The deployment of such AI systems raises questions about cost, vendor lock-in, and governance. Implementing this technology on a large scale would require significant investment in infrastructure and training. Additionally, concerns about data privacy and the potential for misuse must be addressed. The market reaction to this innovation will depend on how effectively these challenges are managed and how the technology is integrated into existing security frameworks.
While the system is still in the experimental phase, its potential applications in security and law enforcement are significant. The research team has indicated that further development is needed to improve the system's accuracy and expand its capabilities. As the technology evolves, it may become a valuable tool for detecting concealed weapons in public spaces, airports, and other high-security areas.