GNSS/PNT is the satellite positioning, navigation and timing layer that almost all modern infrastructure silently depends on, and the current research frontier is not better satellites but detecting, attributing and surviving deliberate jamming and spoofing of it.
Summary
Global navigation satellite systems (GPS, Galileo, GLONASS, BeiDou and their augmentations) deliver position and time by broadcasting extremely weak ranging signals from medium Earth orbit. A receiver correlates those signals, solves for its own position and clock offset, and in doing so becomes a source of both navigation and of precise time. Because the signals arrive at power levels near the thermal noise floor, they are trivially overpowered by a local transmitter (jamming) or replaced by a plausible counterfeit (spoofing). That physical asymmetry, rather than any deficiency in the space segment, is what defines the technology’s present problem set.
The body of work now forming around GNSS/PNT is therefore mostly a resilience layer. It has four strands. First, detection and classification of interference at or near the receiver: compressing raw GNSS streams and classifying jamming or spoofing in real time on edge hardware such as a Google Edge TPU using variational autoencoders, or folding GNSS jamming classification into a multi-task RF recognition accelerator that reports 99.5% accuracy on a GNSS jamming dataset with 98 microsecond per-frame latency. Second, geolocation of emitters: reinforcement-learning agents that actively move a 2x2 patch antenna to disambiguate multipath-corrupted snapshots, and received-power plus time-difference-of-arrival methods applied to reference-station networks. Third, space-based monitoring: using GNSS reflectometry delay-Doppler maps from the CYGNSS constellation to spot terrestrial RFI, where constellation size drives detection latency and coverage. Fourth, architectural defences and alternative PNT, including anchor-rooted recovery for spoofed UAV swarms and terrestrial LPWAN time-difference-of-arrival positioning for cases where GNSS is ruled out by energy cost or indoor operation.
The parameters that decide outcomes are: how much interference power and geographic reach an adversary can field (a space-based interferer changes the scale entirely, and one has been identified as a constellation of Russian early warning satellites in Molniya orbits causing wide-area events over Europe, Greenland and Canada since 2019); how many sensors a monitoring network has, since gains from adding satellites are steepest between one and three; the power and cost budget for putting detection into receivers, which is why ultra-low-power GNSS silicon matters commercially ref; and whether a defence has access to an absolute positional reference rather than only relative geometry, which is provably insufficient against a rigid translation.
Viability (unscored)
Drivers (unscored)
Novelty (unscored)
Diffusion (unscored)
Impact (unscored)
Timing Unclear
Assessment drafted 2026-08-31 from up to 11 KB sources using the technology-scorecard framework; scores are a draft read pending review.