This paper presents the performance analysis of signals from the Galileo satellites in the E1 and E5a frequency bands and GPS L5 signals as measured by DLR’s experimental ground-based augmentation system. The results show that the raw noise and multipath level of Galileo signals and of the GPS L5 signals are smaller than that of GPS L1. The new signals are also less sensitive to the choice of carrier-smoothing time constant. Furthermore, the inter-frequency biases that affect dual-frequency processing are investigated. These biases differ between satellites and depend on satellite and receiver hardware, but they can be determined a priori. With known receiver and antenna configurations, it is possible to correct for these biases. A residual uncertainty associated with the bias correction has to be taken into account. This can be modeled as part of the ground and airborne bounding standard deviations (spr_gnd and spr_air) used in GBAS processing.

/pdf/evaluation-of-gps-l5-and-galileo-e1-and-e5a-performance-for-5ao16zetu1.pdf

Evaluation of GPS L5 and Galileo E1 and E5a Performance for Future Multifrequency and Multiconstellation GBAS

Antenna array processing techniques are studied in GNSS as effective tools to mitigate interference in spatial and spatiotemporal domains. However, without specific considerations, the array processing results in biases and distortions in the cross-ambiguity function (CAF) of the ranging codes. In space-time processing (STP) the CAF misshaping can happen due to the combined effect of space-time processing and the unintentional signal attenuation by filtering. This paper focuses on characterizing these degradations for different controlled signal scenarios and for live data from an antenna array. The antenna array simulation method introduced in this paper enables one to perform accurate analyses in the field of STP. The effects of relative placement of the interference source with respect to the desired signal direction are shown using overall measurement errors and profile of the signal strength. Analyses of contributions from each source of distortion are conducted individually and collectively. Effects of distortions on GNSS pseudorange errors and position errors are compared for blind, semi-distortionless, and distortionless beamforming methods. The results from characterization can be useful for designing low distortion filters that are especially important for high accuracy GNSS applications in challenging environments.

/pdf/assessment-of-measurement-distortions-in-gnss-antenna-array-5b89t20pwq.pdf

Assessment of Measurement Distortions in GNSS Antenna Array Space-Time Processing

Intelligent transport system applications use the global navigation satellite systems (GNSSs) to deliver the information on position, velocity, and time. With the expected deployment of critical road services, GNSS-based positioning terminals will have to be sufficiently accurate and their service will have to be available under any environmental condition. As the answer to the absence of standards and certification procedures of positioning terminals in road transport services, the European COST Action SaPPART was launched and the European norm EN 16803-1 was accepted. They provide the methodology and the guidelines for the terminal performance assessment. Section I presents the data collection methodology based on European norm EN 16803-1 with the definition of performance metrics and classification, operational scenarios, standardized GNSS environments, and field test guidelines. It is followed by the description of the measurement setup, including the trajectories, experimental vehicle with a reference trajectory system, and 17 tested GNSS receivers. The performance assessment of geodetic, automotive, and smartphone GNSS receivers was carried out through the extensive data collection campaign in Nantes, France. The campaign covered four types of environments: clear-sky, rural, peri-urban, and urban. The presented results focus on the basic performance features of accuracy, availability, and the classification of positioning terminals according to SaPPART methodology. The final conclusion is that for achieving a sub-meter accuracy at 95% confidence level in mixed road scenarios, a hybrid multi-sensor positioning approach is required for the majority of mass-market receivers.

Positioning Performance Assessment of Geodetic, Automotive, and Smartphone GNSS Receivers in Standardized Road Scenarios

The signal-in-space of the first GPS Block III spacecraft is analyzed based on radio frequency measurements collected with a 30-m high-gain dish antenna as well as data from geodetic GPS receivers. The spectral properties and modulation characteristics are discussed with focus on the L1 band, which employs a novel interlaced majority voting technique for combination of the C/A, P(Y), and L1C data + pilot signal components. Compared to the preceding generation of Block IIF satellites, a modified shaping of the L1 transmit antenna gain pattern is found, which results in lower carrier-to-noise density ratios at mid to high elevations. Along with this, use of a separate transmission chain for the military M-code signal is evidenced through the analysis of in-phase/quadrature signal components and the derived transmit antenna gain variations. A high level of signal purity is demonstrated on all frequencies, which can be attributed to the use of a new, mostly digital, signal generation unit. Maps of code bias variations for selected signals are presented to quantify the achievable user tracking performance as a function of user receiver parameters. For the L5 signal, a notable reduction in digital distortions is obtained with respect to the Block IIF satellites, whereas analog distortions are found to be of similar magnitude. Thermally induced L5 phase variations found in the Block IIF satellites are no longer observed in GPS III. Using triple-frequency phase observations, a sub-centimeter consistency of the L1, L2, and L5 carriers is demonstrated.

/pdf/signal-analysis-of-the-first-gps-iii-satellite-2ojeghiqxp.pdf

Signal analysis of the first GPS III satellite

Satellite-based augmentation systems (SBAS s) are designed to enhance the performance of standard global navigation satellite system (GNSS ) positioning. SBASs improve the positioning accuracy by providing corrections for the largest error sources. More importantly, SBASs provide assured confidence bounds on these corrections that allows users to place integrity limits on their position errors. Several systems have been implemented around the world and several more are in development. They have been put into place by civil aviation authorities for the express purpose of enhancing air navigation services. However, SBAS services have been widely adopted by other user communities, as the signals are free of charge and easily integrated into GNSS receivers.

Satellite Based Augmentation Systems

Modeling both nominal distortions and signal anomalies (i.e. Evil waveforms, International Civil Aviation Organization (ICAO) threat models) is crucial for an accurate user error budget for Space-based Augmentation Systems (SBAS) and Ground-based Augmentation systems (GBAS) (multi-frequency, multi-constellation). Such analysis is of particular interest for the ongoing standardization work of L1-L5 SBAS services. In this paper we propose a method to jointly estimate the analog and the digital distortions. This comes very useful in those cases in which the estimates of the two satellite imperfections are interdependent, e.g. when the time-bandwidth product is small. The impact of both satellite imperfections on the ranging performance of a Global Navigation Satellite Systems (GNSS) receiver is assessed by means of analytical formulas which make use of the cross-power spectral density (CPS) between the distorted and the ideal signal, and of the power spectral density (PSD) of the nominal (ideal) signal.

Characterization of Nominal Signal Distortions and Impact on Receiver Performance for GPS (IIF) L1/L5 and Galileo (IOV) E1 /E5a Signals

The scenario of GNSS is speeding up on a no-return track
towards a multi-constellation and multi-frequency GNSS
world. On top of the good-ole-boy GPS, the re-born
GLONASS system is steadily improving. The exciting
perspectives of new global systems as Galileo or Compass
are getting closer to reality. Regional systems as QZSS or
IRNSS are also on the starting line. Augmentation
systems flourish worldwide: WAAS, EGNOS, MSAS,
GAGAN…
Within this seemingly chaotic scenario of possibilities,
some initiatives are actively working on the definition of
efficient, pragmatic and user-oriented approaches to take
advantage of this new challenging scenario into a
maximum-benefit scheme for all GNSS users.
In particular Europe has launched, under ESA’s European
GNSS Evolution Programme, the MRS initiative. MRS
standing for Multi-constellation/multi-frequency Regional
System, the initiative is putting different teams of experts
into the exciting goal of defining the paths for the most
successful GNSS usage on a future of multiple choices.
Following this approach and based on the promising study
results obtained so far by ESA (see ref [1]) MRS is
increasingly becoming a serious and natural candidate for
the evolution of the current EGNOS System which is
qualified and in operations since 3 years and which
delivers both an Open Service and a Safety Of Life
aeronautical service (same as WAAS). EGNOS is on the
way to certification planned for mid 2010.
But MRS is not just looking far in the wonderful future of
several matured GNSS systems. Also the present and
short term future are being taken into account, based on
what we have today and aiming to improve the services
that we are already providing to ourselves today, as GNSS
users.
Today we do have already a multi-constellation scenario,
with GLONASS getting close to having the nominal
constellation in operation. Furthermore, processing
platforms as that of EGNOS, magicSBAS or SPEED [2]
data into an SBAS solution.
Based on these available tools, ESA has prepared a
detailed experimentation plan aimed to demonstrate the
benefits that the multi-constellation approach can provide
to today’s GNSS users. This plan covers a wide range of
objectives covering different user domains (aeronautical,
land mobile, maritime) equipped with different types of
receivers. This set of experimentations will rely on the
deployment of several Test Beds combining multi
constellation multi-frequency processing and multi
broadcast channels (GEO, MEO, terrestrial). Among the
first experimentations planned, there are some devoted to
experiment and validate the performance benefits for
SBAS users (GPS L1 only first then GPS/GLONASS or
GPS/Galileo – L1/E1 mono frequency or dual frequency
E1/E5 SBAS users).
This paper presents the outcome of the first step of this
experimentation campaign, which has been performed
based on the magicSBAS tool, a flexible SBAS
processing platform, able to acquire single and dual
frequency GLONASS data, in addition to GPS, in order to
compute and provide both standards SBAS corrections
and integrity, as well as augmentation to GLONASS.
The objectives of this first step are twofold:
1. Analyse the benefits for single frequency users of
using both GPS and GLONASS augmentation, in
particular for uncontrolled environments, where the
visibility of additional satellites play a key role on
the reliability of the GNSS services, as for instance
for urban applications.
2. Analyse the potential benefits that can be achieved
by improving the standards GPS L1 SBAS
augmentation service, when GLONASS data are
used within the processing facility to improve the
overall accuracy, integrity and continuity of the
service.
In particular we will analyse how such a dualfrequency/
dual constellation processing can improve
ionospheric information, by increasing:
- accuracy, with more visible satellites
- robustness, for instance against scintillation effects
- service coverage, by offering improved (more
accurate and more available) Ionospheric Grid Point
monitoring to users located at the service area
boarder;
- reliability with better geometry for integrity
monitorisation
The results of the analysis will be used as a key proof of
concept for the definition of the short term evolutions of
current European GNSS initiatives, which goes beyond
the usage of GLONASS towards the incorporation of
Galileo in a truly System of Systems, where all users,
including legacy users will be able to exploit the
advantages stemming from optimized performances,
higher reliability, maximum integrity and seamless
navigation.
MRS is one of several worldwide initiatives, but it is
marching steadily and actively towards its goals, and
there is no better way to march than by experimenting,
trying and learning from practice.

The Benefits of Multi-constellation GNSS: Reaching up Even to Single Constellation GNSS users

Safety critical applications within the fields of aviation and ground transportation using Global Navigation Satellite Systems (GNSS) signals are present or in the stage of certification in these days. In this context the effect of nominal signal distortions on user error has to be assessed. 

Actually the GNSS signals radiated by satellites do not match perfect waveforms. This is due to several non-idealities of the satellite payloads; Some of them are observed under payload nominal conditions which are called ”nominal distortions”, and others encountered under failure conditions are called ”anomalous signal deformations” (or evil waveforms). It is of importance for several GNSS applications (e.g. civil aviation) to carefully assess the undesired effects of nominal distortions for each satellite of a given GNSS constellation, as different satellites introduce different distortions [1].

In this work we use a method for jointly estimating analog and digital (nominal) distortions presented in [4,5]. In addition to previous work we analyse the nominal distortions over time. Measurements of DLR’s 30m high gain antenna are used to derive estimates of nominal distortions for different GNSS satellites over the duration of the snapshot (short term) and the full path. This will provide first insight into the stability of nominal signal deformations. In this work we do not restrict ourselves to modelling linear analog distortions by a specific characteristic, e.g. a second order system. The reason for that is that the second order system model is not always accurate for legacy L1 signals as other previous studies have shown [2]. Moreover, as will be shown in the final paper, especially for signals with lower time-bandwidth product as e.g. the new GPS L5 or Galileo E5a signals the second order system model turns out not to provide a satisfactory modelling accuracy. We will show validation methods of the signal modelling versus real signal measurements to ensure the usability of the presented results and proposed options for future modelling strategies to improve the accuracy of future signal distortion analyses. 

Furthermore, these estimates of analog distortions as well as the related models are input to a GNSS receiver performance model following the ones given in [3] and [4] with which the tracking error bias and variance are derived with respect to discriminator types, receiver bandwidth, correlator spacing, etc. This will provide means to assess nominal user error behavior and also derive assumptions for nominal receiver errors for different GNSS satellites. We also will provide means to select receiver parameters in order to achieve small differential user error bias. 

In addition to previous work, we will also consider the derivation of user error bias for an example of satellite constellations in order to study the effect of the derived user error biases in positioning domain. 
The above described work supports European standardization activities and was partially performed within the activity Single European Sky ATM Research Programme (SESAR) funded by the European Commission and EUROCONTROL with respect to Multi-GNSS CAT II/III GBAS (Ground Based Augmentation Systems). This work is also targeting to contribute to the definition of the future SBAS (Space Based Augmentation Systems) L1/L5 multi-constellation user avionics standard, which is an evolution of the current SBAS GPS L1 SBAS standard.


[1] R. Eric Phelts and Dennis M. Akos, “Effects of signal deformations on modernized GNSS signals,” Journal of Global Positioning Systems, vol. 5, no. 1, pp. 2–10, 2006.

[2] R. E. Phelts, Todd Walter, and Per Enge,“Characterizing Nominal Analog Signal Deformation on GNSS Signals,” in Proceedings of the 22nd International Technical Meeting of The Satellite Division of the Institute of Navigation (ITM 2009), Savannah, GA, September, 2009.

[3] M. Vergara, F. Antreich, and M. Meurer, “Effect of Multipath on Code-Tracking Error Jitter of a Delay Locked Loop,” in Fourth European Workshop on GNSS signals and signal processing, GNSS Signal 2009, Wessling, Germany, December, 2009.

[4] M. Vergara, M. Sgammini, Y. Zhu, S. Thoelert, and F. Antreich, “Tracking Error Modeling in Presence of Satellite Imperfections”, Proceedings of the ION International Technical Meeting (ITM) 2014, January, San Diego, CA, U.S.A., 2014

[5] S. Thoelert, M. Vergara, C. Enneking, F. Antreich, M. Meurer, D. Brocard, C. Rodriguez, “Characterization of Nominal Signal Distortions and Impact on Receiver Performance for GPS (IIF) L5 and Galileo (IOV) E1 /E5a Signals”, ION GNSS+, 10.-12. Sep. 2014, Tampa, Florida, USA

GNSS Nominal Signal Distortions - Estimation, Validation and Impact on Receiver Performance

Receiver Autonomous Integrity Monitoring (RAIM) has been certified to provide lateral guidance in flight operations ranging from En-route to Non-Precision Approach (NPA). Recent developments in the RAIM algorithm science, namely Advanced RAIM (ARAIM), have suggested a future role in vertically guided operations down to LPV with a decision height of 200ft [EU-U.S., 2012]. However, more stringent requirements as a result of the vertical guidance application question the external risk or trust that is placed on the constellation service provision and may require the partial reduction of this risk through the use of a ground segment, identifying and removing threats and providing data through an ISM (Integrity Support Message). This ground segment should ideally be light and low-cost so not to replicate that implemented for SBAS. In addition the ISM latency [Walter et al, 2012] should ideally be allowed as long as possible to obviate the challenging and expensive communications requirements as imposed, for example, on SBAS (6 sec Time to Alert). Furthermore, the ISM should be as simple as possible to ensure the data broadcast requirements can be met with a number of solutions from ATC, to local ground communications to GEO relay. Finally, the network should be light, in the sense of a sparse and global distribution of stations. In order to meet the defined role of the ground segment and its monitoring capability; three possible methodologies were identified [Milner et al, 2014]: No ground monitoring, Offline Monitoring, Real-Time Monitoring. The parameters of interest to this monitoring are the input parameters defined for the ARAIM baseline airborne algorithm as given below: • URA/SISA: Standard deviation of ranging measurement for integrity • URE/SISE: Standard deviation of ranging measurement for nominal accuracy/continuity • Bnom: Maximum nominal bias on ranging measurement • Psat: Prior probability of fault in satellite per approach • Pconst : Prior probability of fault affecting more than one satellite in constellation per approach This list of parameters contains the maximum nominal bias Bnom. A nominal bias is a fault-free bias, both to account for near-constant uncorrected errors (signal deformation and antenna bias) and non-Gaussian behaviour. However, some small nominal bias may be included in this Bnom parameter. Indeed, after application of all possible corrections, ionofree smoothed code ranges are affected by residual ephemeris plus satellite clock and payload group delay errors w.r.t constellation reference frame and clock. In the context of ARAIM, the residual ephemeris plus clock errors, residual tropospheric error, and multipath plus noise errors, are all assumed to be random errors overbounded by zero mean gaussian errors with known modeled variance. However, it is noted that the residual ephemeris plus satellite clock errors may include a long term bias. These ionofree smoothed code ranges are also affected by the receiver clock offset, defined as the common propagation delays from antenna to signal processing stages, also defined as the error identical to all measurements of the same constellation, which varies across constellations (time reference, signal) and the receiver design. Note that the receiver clock offset may include residual payload plus ephemeris delays identical to all satellites used in the navigation solution computation, so may vary depending on the set of satellites used in this computation. The iono-free nominal bias may then be defined as the permanent bias in excess of the residual error identical to all measurements of the same constellation, and from this definition may therefore depend on the receiver clock offset. A first paper has been issued to define properly the nominal bias and to characterize over the globe those biases for an ARAIM user [Macabiau et al, 2014]. Three possible types of sources of nominal bias were identified: nominal signal deformation, variation of SV antenna group delay with nadir angle, variation of user antenna group delay with Azimuth (Az) and Elevation (El) angles. Models used to characterize theses nominal bias contributions were proposed and fully defined. Assumptions were made at several levels of these models to try and reflect possible nominal situations of signal distortion, SV or user receiver antenna group delay variation. Initial work was presented on the ARAIM reference algorithm integrity monitoring performance to protect the ARAIM user against the impact of these nominal biases, driven by the ISM input Bnom value transmitted by the ground segment. The aim of this paper is therefore to update the analysis done on nominal bias affecting the ARAIM user, on the capacity of the ground monitoring network to provide a pertinent Bnom, and on the impact on the ARAIM user performance. This methodology allows determining possible restrictions on ARAIM user receiver characteristics. Based on the definition of the ARAIM user nominal bias expressed in [Macabiau et al, 2014] identifyng three possible sources of nominal bias (signal deformations, SV antenna, and user antenna), assumption and models definition are set in the first part of the paper. Then, we determine the impact of that defined nominal bias on the ARAIM user receiver range measurement and position estimate. Impact of nominal signal deformations is evaluated using a models derived from the bounding ICAO EWF threat model The evaluation considers different receiver configurations in terms of bandwidth and chip spacing, representing the regions that are proposed at RTCA/EUROCAE [Phelts et al., 2014b] plus regions identified to induce a maximum ranging error due to nominal signal deformation. The analysis of the nominal bias obtained for the different configuration leads to the identification of suitable design requirements for the ARAIM user receiver. Impact of user antenna group delay variation as a function of Azimuth and Elevation is then addressed, considering several models for user antenna, including recent results for the model of a dualfrequency civil aviation antenna mounted on aircraft. Impact of SV antenna group delay variation is also assessed based on mathematical analysis of antenna group delay. Then, through simulation, we analyze the results of the implementation of these ground monitoring techniques for estimation of the Bnom bounds and we analyze the performance of these bounds with respect to the possible distribution of the ARAIM user nominal bias. Situations leading to extreme integrity situations have been identified and are analyzed with respect to current monitoring concept used in the ISM. This analysis will assess the sensitivity of the results with respect to the model used to define each nominal bias contributor. From these simulations, we finally conclude on the capacity of the ground to provide efficient Bnom coupled with some possible restrictions for the characteristics of the ARAIM user receiver.

Nominal Bias Analysis for ARAIM User

Receiver Autonomous Integrity Monitoring (RAIM) has been certified to provide lateral guidance in flight operations ranging from En-route to Non-Precision Approach (NPA). Recent developments in the RAIM algorithm science, namely Advanced RAIM (ARAIM), have suggested a future role in vertically guided operations down to LPV with a decision height of 200ft. However, more stringent requirements as a result of the vertical guidance application question the external risk or trust that is placed on the constellation service provision and may require the partial reduction of this risk through the use of a ground segment, identifying and removing threats and providing data through an ISM (Integrity Support Message). This ground segment should ideally be light and low-cost so not to replicate that implemented for SBAS. In addition the ISM latency should ideally be allowed as long as possible to obviate the challenging and expensive communications requirements as imposed, for example, on SBAS (6 sec Time to Alert). Furthermore, the ISM should be as simple as possible to ensure the data broadcast requirements can be met with a number of solutions from ATC, from local ground communications to GEO relay. Finally, the network should be light, in the sense of a sparse and global distribution of stations.

In order to meet the defined role of the ground segment and its monitoring capability; three possible methodologies were identified: No ground monitoring, Offline Monitoring, Real-Time Monitoring.
The parameters of interest to this monitoring are the input parameters defined for the ARAIM baseline airborne algorithm as given below:
•	URA/SISA: Standard deviation of ranging measurement for integrity
•	URE/SISE: Standard deviation of ranging measurement for nominal accuracy/continuity
•	Bnom: Maximum nominal bias on ranging measurement
•	Psat: Prior probability of fault in satellite per approach
•	Pconst	: Prior probability of fault affecting more than one satellite in constellation per approach 
This list of parameters contains the maximum nominal bias Bnom. A nominal bias is a fault-free bias, both to account for near-constant uncorrected errors (signal deformation and antenna bias) and non-Gaussian behaviour. However, some small bias may be included in this Bnom parameter.

Indeed, after application of all possible corrections, iono-free smoothed code ranges are affected by residual ephemeris plus satellite clock and payload group delay error w.r.t constellation reference frame and clock. The residual ephemeris plus clock error (URE), residual tropospheric error, and multipath plus noise error, are all assumed to be random errors overbounded by zero mean gaussian errors with known modeled variance. However, note that the residual ephemeris plus satellite clock error may include a permanent bias, labeled here nominal bias.

These ionofree smoothed code ranges are also affected by the receiver clock offset, defined as the common propagation delays from antenna to signal processing stages, also defined as the error identical to all measurements of the same constellation, which varies across constellations (time reference, signal) and the receiver design. Note that the receiver clock offset may include residual payload plus ephemeris delays identical to all satellites used in the position computation, so may vary depending on the set of satellites used in this computation.

The iono-free nominal bias may then be defined as the permanent bias in excess of the residual error identical to all measurements of the same constellation, and from this definition may therefore depend on the receiver clock offset.

Proper overbounding of this nominal bias by a ground segment to be sent via an ISM may therefore be difficult depending on the size of unmodeled values. On the ARAIM user side, the nominal bias affects the measurements but is not used in the MHSS detection threshold determination, although it is partially reflected with the URE. So the problem is that bounds on the nominal bias are to be known by the ARAIM user receiver, and impacts ARAIM user integrity and availability. Therefore, Bnom bounds on biases from ISM shall cover with high probability all nominal biases for all users.

On the ARAIM ground segment side, it is anticipated that Bnom is determined from integrity analyses and accounts for nominal signal deformation and antenna biases as well as other potentially non-zero mean error contributions. But it is felt that clarification on the definition of these nominal biases, analysis on potential techniques to be used on the ARAIM ground segment, determination of their performance, and final impact on ARAIM user is needed.

The aim of this paper is therefore to determine the size of the nominal biases affecting the user measurements, the capacity of the ground monitoring network to provide a pertinent Bnom, and the impact on the ARAIM user performance. This methodology could allow determining possible restrictions on ARAIM user receiver characteristics.

C. Rodriguez

Papers

Characterization of Nominal Signal Distortions and Impact on Receiver Performance for GPS (IIF) L1/L5 and Galileo (IOV) E1 /E5a Signals

The Benefits of Multi-constellation GNSS: Reaching up Even to Single Constellation GNSS users

GNSS Nominal Signal Distortions - Estimation, Validation and Impact on Receiver Performance

Nominal Bias Analysis for ARAIM User

Impact of ARAIM Nominal Bias Bounding Techniques on Final ARAIM User Performance