Richard Stottler, Stottler Henke Associates Inc, James Ong, Stottler Henke Associates Inc, Chris Gioia, Stottler Henke Associates Inc, Chris Bowman, Data Fusion & Neural Networks, Apoorva Bhopale, Air Force Research Lab
Keywords: Electro Magnetic Interference (EMI), Terrestrial EMI, EMI detection and Identification, Artificial Intelligence, Neural Networks, Behavior Transition Network (BTN)
Abstract:
Clear operating spectrum at ground station antenna locations is critically important for communicating with, commanding, controlling, and maintaining the health of satellites. Electro Magnetic Interference (EMI) can interfere with these communications, so it is extremely important to track down and eliminate sources of EMI. The Terrestrial RFI-locating Automation with CasE based Reasoning (TRACER) system is being implemented to automate terrestrial EMI emitter localization and identification to improve space situational awareness, reduce manpower requirements, dramatically shorten EMI response time, enable the system to evolve without programmer involvement, and support adversarial scenarios such as jamming. The operational version of TRACER is being implemented and applied with real data (power versus frequency over time) for both satellite communication antennas and sweeping Direction Finding (DF) antennas located near them. This paper presents the design and initial implementation of TRACER’s investigation data management, automation, and data visualization capabilities.
Date of Conference: September 19-22, 2017
Track: Poster