Standards Development

Scope

The accuracy performance of a GNSS receiver is influenced by a set of technical and environmental parameters.

It depends first on the characteristics of GNSS infrastructures (satellite constellations, orbital and clock stabilities, atmospheric propagation, etc.) and on the availability of assistance signals such as differential corrections, SBAS, RTK, or PPP, which directly contribute to the reduction of positioning errors.

Performance is also affected by various disturbances. “Passive disturbances” originate from local propagation phenomena, including diffraction, reflections (multipath), and masking. “Active disturbances” correspond to intentional or unintentional radio frequency unexpected signals, including jamming, spoofing, and meaconing. To these must be added disturbances related to “environmental conditions”, such as extreme temperatures, shocks, or vibrations, which may also degrade the stability of the processed signal.

Finally, accuracy is also sensitive to rapid variations in dynamics and attitude, such as accelerations, decelerations, sharp turns, or U-turns. These abrupt changes challenge navigation algorithms, generating transient errors.

For the evaluation to be representative, GNSS metrological assessment of accuracy must therefore simultaneously integrate these three dimensions.

In the GNSS domain, errors affecting receivers are generally divided into two main categories: 

• Systematic errors, which manifest as similar biases observed across a set of GNSS receivers of the same model, leading to recurring deviations with respect to the true trajectory followed by the vehicle.

• Random errors, which are unpredictable and can only be described through a statistical uncertainty envelope. These random errors represent most disturbances encountered in terrestrial environments (urban, agricultural, road, rail, river and in flight for drones).

Test methods thus consist in multiplying measurements for a given scenario: 

• either by simultaneously embedding several receiver models to be evaluated,

• or, in a more controlled manner, by replaying GNSS signals previously recorded on site, corresponding to the targeted environments and scenarios.

The test data are then consolidated into a sequence of vectorized points, which are analysed statistically. This so-called “trajectory set” approach provides reliable and accurate results, provided that the method is correctly applied from planning through test data collection to trajectory analysis, with an instrumentation properly managed and sufficiently performant.

The latter method, known as the replaying technique, is defined in the two complementary European standards EN 16803-4 and EN 16803-2, which respectively describe the procedures for recording and replaying GNSS signals collected in real-world environments.

The purpose of this initiative is to establish an international standard specifying the minimum requirements necessary to ensure reliable, accurate, and reproducible results for the performance assessment and classification of GBPTs.

In a first part, to ensure a representative evaluation, the future standard will require that data collected in real-world environments comply with strict conditions regarding traceability and synchronization of test measurements. The instrumentation chain shall be properly calibrated to quantify the uncertainties associated with the reference trajectories and to ensure the faithful digitization of GNSS signals. Data integrity checks, together with comprehensive documentation of test conditions, shall ensure that the recorded datasets can be reliably reused for comparative analyses, metrological validations, or, where applicable, certification procedures.

In a second part, the future standard shall list the minimum requirements for replaying recorded GNSS signals on appropriate test benches. It will verify the proper implementation of the methodology aimed at reproducing the recorded GNSS signals with the highest possible fidelity. The number of iterations shall remain consistent with the expected level of accuracy.

Finally, the data processing and statistical analyses produced shall reflect the behaviour of the GBPT, enabling classification and comparison with subsequent measurement campaigns.

 Together, these two parts shall define a coherent metrological framework that ensures continuity between field data collections and laboratory replay testing, under representative, repeatable, reproducible and affordable conditions.

This test methodology will also be extended to the integration of other disturbances, such as those referred to as active (part 3) or environmental.

Purpose

Approaches based on GNSS simulators or on field test campaigns in operational environments present significant limitations: simulators do not faithfully reproduce the diversity of real-world environments, while on-site testing suffers from a lack of repeatability and reproducibility of results, due to the constantly changing conditions of radiofrequency propagation, starting with the permanent variations in satellite constellation geometry. As a result, these methods do not meet metrological accreditation requirements and cannot serve as a reliable basis for certification.

The purpose of this proposal is therefore to control the proper implementation of multi-measurement test methods aimed at assessing GNSS receiver positioning accuracy performance, under conditions consistent with the expected use cases, in order to obtain exploitable results.

This approach can be implemented either by simultaneously testing several receivers from a same model under identical conditions, or by replaying, on a test bench, GNSS signals previously recorded in representative environments. In particular, the so-called “GNSS signal replay method” ensures results that are reliable, representative, repeatable, and reproducible, compliant with accreditation rules, and therefore suitable for certification.

The deliverable shall provide verification of proper practices regarding test planning, data collection, signal replay (where applicable), and statistical analysis of trajectory sets obtained from multiple measurements. For example:

• Selection of antennas

• Choice of cables and verification of their lengths

• Signal quantization level and gain control management

• Measurement uncertainty as a function of the number of trajectories

• Identification and justification of discarded trajectories

• Use and presentation of results

Standardizing this approach is essential to ensure comparability of results and to facilitate certification processes across a variety of applications.

Furthermore, this methodology can be extended to other categories of disturbances that may add to those from the local environment of a given test campaign, such as GNSS attacks (jamming, spoofing) or extreme environmental conditions (temperature, vibration, shock).