Approximating Regional GNSS Interference Sources as a Convex Optimization Problem Using ADS-B Data

Abstract

While the Automatic Dependent Surveillance-Broadcast (ADS-B) has been widely used for air traffic operations and management, it has also been useful recently in identifying, detecting, and localizing (IDL) potential GNSS/RFI jamming sources in regional, high air traffic environments. With an increase in reported GNSS interference around the world, there is a necessity to find and remove jammers from the environment to prevent additional unsafe air travel operations. The major indicator that infers as whether an aircraft is likely being jammed (from ADS-B) is by monitoring the Navigational Integrity Category (NIC) value included in the ADS-B message. While not as effective as other metrics typically used in interference detection, it can still provide an indication if jamming is present, but presents an opportunity in localizing the potential source in real time. This paper seeks to approximate the area of potential GNSS/RFI interference by fitting a Euclidean Cone to ADS-B data reporting low NIC values. This problem is formulated as a convex optimization problem, which is derived from an alternative version of maximum inscribed ellipsoid approach. By fitting the optimal cone to data potentially impacted by interference, the apex of the cone will reveal the estimated jamming location. Raw ADS-B data is processed, decoded, interpolated and filtered to improve localization results. The proposed convex formulation is then applied to two reported interference events, the first over a period of 36 hours near Denver International Airport in January 2022, and the second over roughly 8 hours near the Dallas-Fort Worth Area in October 2022. The Denver localization results show that the four estimated jamming locations - calculated from four six-hour time windows - are grouped in between the downtown Denver area and the airport. With the Dallas interference jamming localization results, it can be seen that the three estimated jamming locations - determined from three, one hour windows - also show a tighter grouping on the southern side of the Dallas/Fort Worth area.