A Metadata Management and Visualization System for Radio Frequency Activity Modeling and Pattern Recognition

In current Naval Communications Intelligence operations, significant volumes of potentially valuable, “non-analyzable”, intercepted data are discarded. By using an Automated Radio Frequency Activity Modeling and Pattern Recognition (RF-AMPR) system, operators may be able to gain critical insights into RF activities from this discarded data. This capability would be a key enabler for Naval Intelligence. It allows radio fault detection, spectrum interference monitoring, dynamic RF pattern recognition, clustering and tracking of emitters, and anomaly detection. The main objective of this effort is to develop technical underpinnings and methodologies, as well as a workable prototype, for RF activity modeling and pattern recognition to address the challenging problems in the dynamic and dense RF environments. Specific technical objectives of the proposed metadata analysis, management, and visualization system (MDAMS) include: develop a generative model for the RF activities to generate metadata; design a metadata management system to efficiently retrieve, organize, and display the RF activity; cluster detections into contacts for emitters localization and tracking; develop a pattern learning and recognition module for anomaly detection; develop a friendly operator interface for situation monitoring; and demonstrate the performance of the proposed system with realistic and pressing scenarios using simulated and real metadata.