Title: Crowdsourcing GNSS Jammer Detection and Localization
Author(s): Luka Strizic, Dennis M. Akos, Sherman Lo
Published in: Proceedings of the 2018 International Technical Meeting of The Institute of Navigation
January 29 - 1, 2018
Hyatt Regency Reston
Reston, Virginia
Pages: 626 - 641
Cite this article: Strizic, Luka, Akos, Dennis M., Lo, Sherman, "Crowdsourcing GNSS Jammer Detection and Localization," Proceedings of the 2018 International Technical Meeting of The Institute of Navigation, Reston, Virginia, January 2018, pp. 626-641.
Full Paper: ION Members/Non-Members: 1 Download Credit
Sign In
Abstract: GNSS has wide adoption in critical military and civilian infrastructure, requiring reliable operation. Therefore, its susceptibility to interference can result in anything from a minor inconvenience to a life-threatening affair. In this paper, we present a prototype implementation of the GNSS interference detection and localization, or, so-called, J911 system, which crowdsources measurements of GNSS observables from smartphones. Access to Carrier-to-Noise-density ratio (C/N0) and Automatic Gain Control (AGC) level enables the system to distinguish natural causes of signal degradation from intentional jamming. A mobile application was developed for the Android OS, which records current location, per-GNSS-satellite C/N0 and other available GNSS observables. The collected data is sent to a central server, where it is subject to interference detection assessment and visualization. With a high enough density of smartphones, localization methods can be employed, such as time difference of arrival (TDOA) or power difference of arrival (PDOA). With the help and oversight of the Department of Homeland Security, an exercise to test interference mitigation technologies was conducted in 2017, called JamX 17, where we fielded fifteen smartphones with GPS- and GLONASS-capable hardware. At the time of testing, the Android OS did not support AGC level reporting, so four SiGe GN3S Sampler Software-Defined Radios (SDR) were used instead. Analysis of the collected data shows that rough tracking of the jammer is possible, based on measurements from many phones. When the jammer gets closer to a smartphone, both GPS and GLONASS C/N0 decrease in a similar pattern. When the jammer moves away, C/N0 is restored to nominal levels, but decreases in the next phone that the jammer is closing in on, thus allowing phone-density-limited tracking. The C/N0 measurements are validated by comparison to the AGC measurements from nearby SDRs, confirming that jamming, rather than natural obstruction, took place. In the test site area where the phone density was highest, a TDOA method was conducted and provided a good estimate of the jammer location.