Railway transportation systems require accurate and robust localization for safe operations. In the future, Railway signaling is expected to rely on onboard sensors like Global Navigation Satellite Systems (GNSS) in order to reduce installation and maintenance. GNSS positioning can be however challenging in railway environments. In particular, multipath is one of the largest local errors due to its fast dynamic and complex nature.

This paper proposes to model code multipath by distinguishing between error contribution caused by the antenna installation on the train roof and by the reflections of surrounding objects like e.g., buildings during train operation. Multipath caused by the antenna installation is expected to have similar stochastic properties for a given satellite elevation with respect to the user, while multipath caused by the environment is expected to be variable during the operation.

In this paper we focus on the derivation of a conservative error model of multipath and noise caused by the antenna installation and the vehicle structure surrounding the antenna. To do this, we first isolate multipath and noise from other GNSS errors using the Code-Minus-Carrier method with data collected in open sky scenarios. Second, an overbounding error model is derived. The limitation of modeling with restricted set of real data typically found in practice is discussed in detail and we review methods that ensure the independency of samples to build the final probability distributions. A new approach to create separate data sets is ultimately proposed to derive an overbounding sigma.

The derived models can be used as a reference nominal error models to build the null hypothesis of fault detection algorithms that detects the presence of excessive multipath in dynamic scenarios. Results are obtained for different train installations with real data collected during the EU ERSAT-GCC and RAILGAP projects.

The prospect of GNSS meta-signal tracking promises the synergy of both code reliability and the high-precision of sub-carrier observations. The latter has the advantage, in comparison to carrier-phase observations, of having wavelengths in the order of a few meters instead of cm-level. This realizes the possibility of resolving sub-carrier phase ambiguities without the need of a reference station providing positioning solutions with a sub-meter level accuracy. In the frame of the HANDS-CD project led by IGASPIN GmbH, a synthetic meta-signal observation formed from Galileo E5a and E5b signals using the widelaning concept will be demonstrated in this contribution. This analysis is performed based on a simulated kinematic trajectory. The synthetic meta-signal observations are fed into an Extended Kalman Filter-based positioning engine (M-SiPE-tool), which applies the LAMBDA ambiguity fixing method to resolve the sub-carrier ambiguities. To assess the robustness of the positioning filter against signal impairments, the observations of many Galileo satellites are synthetically contaminated by multipath reflection with different amplitudes. The outcome of the positioning engine exhibits successful sub-carrier ambiguity fixing and provides a sub-decimeter positioning accuracy for a code multipath amplitude of less than 30 meters, or for sub-carrier multipath amplitude of less than 0.5 meters.

This paper addresses the critical issue of unwanted interference in airborne GNSS receivers, crucial for navigational safety. Previous studies often simplified the problem, but this work offers a comprehensive approach, considering factors like Earth's reflective properties, 3D calculations, and distinct radiation patterns. It introduces SPIDEH and OEP graphs to visualize interference distribution along flight paths. Results highlight the significance of physical configuration and distance from interference sources on receiver performance. The algorithm developed can assess interference effects on GNSS receivers and aid in selecting optimal flight paths for minimal interference. This research enhances understanding and management of unintentional interference in airborne navigation systems.

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"

In the context of the Jammertest 2023, a collaborative experiment was carried out by the European Commission Joint Research Centre (JRC), the European Space Operations Centre of ESA, the Norwegian Communication Authority, and the Norwegian Defense Research Establishment (FFI) to explore potential RF interference monitoring in the navigation GNSS band from LEO orbit. The experiment utilizes the ESA OPS-SAT satellite and the possibility of transmitting a custom jamming signal pattern in the context of the Jammertest 2023 event. The objective is to validate the feasibility of detecting and locating ground-generated jamming signals using SDR technology on-board LEO. The insight into the signal structure and location provides a unique chance to assess the performance and limitations of this approach in a real-world scenario. The paper presents the processing of raw RF data collected during the in-flight experiment, including the generation of Frequency Difference Of Arrival (FDOA) observables and emitter geolocation. Despite the constraints posed by onboard resources and mission limitations, this work offers a persuasive proof of concept and suggests new guidelines for implementing this technology on future LEO missions

GNSS-based navigation is commonly used in aircrafts as one source for localisation. As reports of pilots about jamming during flight are becoming more frequent, we performed jamming tests with a parked passenger aircraft to assess the criticality of jamming events. We observed the reaction of the aircrafts GPS‑position display in the cockpit and the ADS-B availability during jamming. We analysed the relative power levels at which the GPS-receivers and also the ADS-B broadcasting show negative effects. Beyond others, we could observe significantly long recovery times of the GPS‑position display and the ADS-B system. The results give unique detailed facts about the vulnerability of one type of commercial passenger aircraft and emphasize how important it is to raise awareness of this vulnerability in aviation safety.

Jammertest is the largest known GNSS jamming, meaconing, and spoofing test event in the world which has an open policy towards both user participation and user communication with no restrictions on the sharing of data or publication of results. The organizers implemented several changes and enhancements within the 2023 test campaign to further broaden the appeal and applicability of the tests for as many demographics of GNSS users as possible. More than 200 participants from 19 nations took part in person from 18-22 September at the test sites along the west coast of the Andøy island. This paper summarizes the design and motivation of the tests and test venue with particular attention to the efforts taken to provide users with precision timing and frequency references independent of the denied and disrupted GNSS signals. Aspects of surveilling and enforcing unintentional emissions, and real time communication and coordination to the large number of distributed participants are also discussed.

Global Navigation Satellite Systems (GNSS) have become indispensable for numerous applications, ranging from the synchronization of critical infrastructure to precise navigation applications. It is imperative to establish reliable monitoring systems that ensure the dependability and security of satellite constellations, given society's increasing reliance on GNSS for accurate positioning, timing, and navigation. This study introduces a cutting-edge GNSS monitoring antenna system which deliver high-performance results.

The antenna system is composed of multiple blocks: a reconfigurable 64 elements dual-frequency (L1/L5) array antenna, filtering stage, amplification and phase shifting stages, signal combiners and a GNSS receiver stage. The proposed antenna is modular, i.e., each component can be controlled independently from the others, allowing easy maintenance and more interestingly, re-configurable antenna specifications. To receive multiple GNSS signals at the same time a custom receiver block has been designed to have 8 RF signal input with each beam created by the antenna connected to a GNSS receiver which is multi-constellations and dual frequency. All the signals are then logged or observed live.

A Raspberry Pi is connected to the receivers to actively control the beams using the publicly available satellites ephemerides (using two-line element files) and changing the phase on each antenna element.

The antenna array is built with 8 panels of 8 elements (2x4 distributions) which can be arranged in multiple shapes. Accordingly, the operator can assemble the panels to have a unique and a very high gain beam or he can choose to create up to 8 sub-arrays, i.e. 8 different beams. These sub-arrays can then be oriented toward different directions to have a full-sky coverage. The shape can also be optimized to achieve a high level of interference and multipath mitigation for GNSS applications. By exploiting the analogue beamforming technique, the beams are steerable which enable the tracking of GNSS satellites.

This paper outlines the methodology and findings of a study aimed at enhancing the performance of the GSHARP PPP algorithm through the application of Machine Learning techniques, specifically targeting the mitigation of multipath effects in GNSS measurements.

The approach uses Machine Learning clustering techniques on key parameters of GNSS measurements to identify and understand patterns in multipath effects. This method serves as an initial step in integrating AI technology into the GSHARP PPP algorithm, with a deliberate choice of lightweight Machine Learning techniques to reduce the integration risk of the outcomes in final implementations. However, more advanced deep learning techniques are not ruled out for future exploration.

Machine Learning algorithms were trained using over 80 hours of data from varied environments like highways, suburban, and urban areas, ensuring their robustness and adaptability in real-world scenarios. The effectiveness of this clustering methodology is evaluated through its integration into the GSHARP PPP algorithm. Different iterations of the PPP algorithm are conducted, incorporating the insights gained from the clustered GNSS measurements in varying ways. The study explores the impact of different clustering strategies on the final positioning solution and the impact of different ways to use the outputs of the clustering on the PPP algorithm.

Preliminary findings suggest a promising enhancement in the robustness and accuracy of GNSS positioning, reducing the maximum errors in several scenarios and showcasing the potential of the introduced Machine Learning techniques. The conclusions drawn from this research not only contribute to advancing GNSS positioning methodologies but also open avenues for future exploration, including the potential integration of more advanced deep learning techniques to further optimize multipath mitigation in high-precision positioning systems.

The vast majority of GNSS users move in urban areas, where the signal conditions are highly unstable and multipath or gross errors make GNSS navigation unreliable or plainly unfeasible. In this study, features from real GNSS data collected by different grades of receivers have been compared, to find candidate statistical indicators of the context that allow the automatic recognition of open sky or obstructed environments. The considered features are all pre-PVT and snapshot-based, hence suitable for real-time applications. They are, namely: the number of visible satellites, the dilution of precision, the multipath linear combination with dual frequency measurements and the C/N0 difference between each couple of satellites in the same epoch at the same frequency. All measurements have been gathered both in open sky and in obstructed scenarios. Evidence suggests the multipath linear combination and the C/N0 difference between couples of satellites as the most promising baselines for an environment classifier based on Machine Learning.