Interference signals can disrupt global navigation satellite system (GNSS) receivers. The
ability to classify the type of interference source is beneficial to counteract and eliminate
an interference source after its detection. For example, interference through legal spectral
coexistence with other radio frequency systems exists [1]. Such interference necessitates a
different reaction than, for instance, an illegally generated jamming signal. The knowledge
about the interference type gained from classification can then guide further steps towards
localization and, finally, depending on the type, elimination.
Previously, a GNSS interference monitoring, detection, and classification system built
around low-cost commercial-off-the-shelf (COTS) sensor hardware utilizing an external server
for interference signal type classification was introduced [2]. This paper extends previous
research by an entirely local interference detection and classification implementation on the
low-cost COTS sensor. The proposed method reduces sensor costs and allows real-time
operation without the need for external processing. The sensor’s capabilities are expanded
by combining conventional statistical signal processing approaches with machine learning for
local quasi-real-time interference signal classification. Features explored in detail in [1] and
extreme gradient-boosted trees are implemented, yielding good classification performance
while staying within the hardware constraints of the low-cost sensor. We show that combining
the detection strategy from [2] with an energy detector enhanced by noise floor compensation
offers improved detection performance while reducing the need for manual calibration. In
addition, we present an end-user-oriented interface enabling non-expert users’ intuitive
operation of the sensor. We conclude with an in-the-field sensor evaluation under real-world
conditions, demonstrating its operational readiness with fully local detection and classification
capabilities.
[1] J. R. Van Der Merwe et al. "Optimal machine learning and signal processing synergies
for low-resource GNSS interference classification"
[2] J. R. Van Der Merwe et al. "Low-Cost COTS GNSS Interference Monitoring, Detection,
and Classification System"