A report on GNSS performance standards
GNSS service providers are to be commended for providing performance standards, and verifying they are being met. With multiple constellations, there is an opportunity to learn from one another and improve the service commitments
Today, the world is experiencing a jump in GNSS/RNSS capabilities and services, advancing beyond the decades-long monopoly of GPS, and to some extent GLONASS, in providing worldwide positioning and timing services. Galileo and BeiDou Navigation Satellite System (BDS) now offer limited global services, and Quasi-Zenith Satellite System (QZSS) and Navigation with Indian Constellation (NavIC) offer regional services.
We are in a significant transition period as new satellite navigation service providers move from initial operation to full operation over the next several years. While each system provides similar services, these services differ in implementation and capabilities, in signals, codes, navigation data, data authentication levels, and ancillary services. Satellite navigation service providers issue documents that describe the service they are providing with commitments on the level of service that users can expect. These include core performance standards such as accuracy, availability, continuity, integrity, and coverage.
In this article, the term “performance standard” is used. Some organizations refer to their performance standards documents by other terms, such as a service standard, open service standard, or service definition document. For the purpose of this article these terms are considered synonymous.
When the United States first offered a global navigation satellite service through its Navstar Global Positioning System (GPS), it became clear to leadership that for its open service to be adopted by users, written commitments of the service were required. This began with a publicly available interface control document, “Navstar GPS Space Segment / Navigation User Interfaces”, ICD-GPS-200, released January 25, 1983 (ICD-GPS-200V1, 1983). While this described the format and content of the open service signal, it did not describe the performance levels that the service would provide.
During the same year GPS reached initial operational capability in 1993, the U.S. Department of Defense issued its first statement of open service, called the Global Positioning System Standard Positioning Service (SPS) Signal Specification (GPSPS1, 1993). Contained in the document was a letter stating that the signal specification defined “GPS services provided by the Department of Defense to the Department of Transportation to support the needs of civil users.” The document included a statement of policy for the provisioning and usage of the system, system overview, and detailed description of ranging signal characteristics. It also included an annex specifying the Standard Positioning Service, including coverage, service availability, service reliability, and positioning & timing. Later versions appeared at subsequent key points. Version 2 was published in 1995 to coincide with full operational capability (GPSPS2, 1995), Version 3 in 2001 to report amended service levels following the turn off of selective availability (GPSPS3, 2001), and Version 4 appeared in 2008 (GPSPS4, 2008). A Version 5 is in development to specify new services, including possibly the L2C and L5 frequencies and codes.
Other satellite navigation service providers also issued their own service standards, and today we have the following:
• BeiDou Navigation Satellite System Open Service Performance Standard, (Version 2.0), December 2018 (BDSOSS2, 2018)
• European GNSS (Galileo) Open Service Service Definition Document, May 2019 (GalOSSDD1, 2019)
• Стандарт эксплуатационных характеристик открытого сервиса (СТЭХОС), translated “Global Navigation Satellite System GLONASS Performance Standard for Open Service”, Revision 2.2, 2019 (GLOOSSP1, 2019)
• Quasi-Zenith Satellite System Performance Standard, November 5, 2018 (QZPS1, 2018)
The value of standards
To support safety-of-life applications, space and ground-based systems have been developed to augment the services provided by the core GNSS services. The Wide Area Augmentation System (WAAS) is one such system that augments the GPS SPS by broadcasting differential correction messages through GEO satellites and supports signal integrity by transmitting integrity data in the WAAS messages, thereby making the augmented GPS Standard Positioning Service a trusted air navigational aid. Putting this another way, WAAS could not offer its service levels without a corresponding commitment from the GPS SPS. When the WAAS performance standard was issued in November 2008, then Federal Aviation Administration GNSS Program Manager Leo Eldredge explained, “For civil aviation purposes, we depend on the commitments contained in the SPS PS as the basis for our commitment to provide service through augmentation and the approvals for aviation use of standalone GPS. Before we could approve use of L5, for either standalone use or as part of an augmented service, the commitment to provide that service would first need to be provided by DoD in a [Performance Standard].” (InsideGNSS, 2008).
The performance standards provided by each of the GNSS and RNSS systems have many similarities, although the commitment levels vary. Table 1 compares the contents of the performance standards for GPS, GLONASS, Galileo, BDS, and QZSS. In some cases, the performance parameters were consistent across most or all of the systems. In other cases, the performance parameters applied only to one of the systems, and these were omitted from the comparison. The table below summarizes the performance level commitment provided in the performance standards for each of the performance parameters. Note that GPS and GLONASS are fully operational, while Galileo, BeiDou, and QZSS are still in development, which influences to a degree their performance commitments.
In reviewing the broad categories of performance parameters, it is observed at this point that some performance parameters are universally provided by all GNSS/RNSS service providers. These include slot/service availability, user range error accuracy, and UTC Offset Error accuracy. Some of the performance parameters are provided by all global though not regional services, including PDOP and positioning service availability. And some performance parameters are not provided until the system’s full implementation, like terrestrial service volume coverage, multifrequency ranging accuracy, and positioning accuracy. There is a desire on the part of GNSS/RNSS service providers to standardize the implementation of their performance standards, which is discussed in the next section.
“Standardizing the standards”
In 2011, the United Nations International Committee on GNSS (ICG) met in Tokyo, Japan, and adopted a recommendation to form a consensus on open service GNSS performance parameters, including definitions and calculation methods (ICGWGA, 2011). In the recommendation, it was proposed that a working group develop a template for individual GNSS providers to consider in their publication of signal and system information, the policies of provision, and the minimum levels of performance to be offered for open services. This work has been carried forward by a group within the Working Group on Systems, Signals, and Services, called the Performance Standards Team, comprised of representatives of China, European Union, India, Japan, Russia, and United States. In 2018, the ICG adopted a set of guidelines for developing performance standards prepared by the Performance Standards Team (ICGPSG, 2018). The Team continues to work on definitions for each of the parameters used in the Guidelines document. Table 2 provides a list of the performance parameters by category that are considered for implementation in GNSS/ RNSS performance standards. While the Guidelines document identifies which performance parameters to include in a performance standard, these are not binding. The ICG operates by consensus, meaning that a position is adopted only if it reflects the views of all its members. Performance parameters that are determined to be critical in defining a performance standard are identified as Key. The other performance parameters are listed as Optional. The expectation is that all members will implement the Key performance parameters and may or may not implement the Optional performance parameters.
Table 2. Guidelines for Performance Standards – Categories and Performance parameters
Slot Availability (maintenance of satellites to orbital slot parameters) [Optional, proposed to be Key]
Terrestrial Service Volume Coverage [Key]
Space Service Volume Coverage [Optional]
Range Accuracy (all signals) [Key]
Range Accuracy (by Age of Data) [Optional]
Range Integrity [Optional]
Range Availability [Key, proposed to be Optional]
Range Rate Accuracy [Optional]
Range Acceleration Accuracy [Optional]
Range Rate Integrity [Optional]
Range Acceleration Integrity [Optional]
This section applies if positioning is provided as a service.
DOP Availability [Key]
Position Accuracy (Global Average & Worst Site)[Optional]
Position Availability [Key]
Time transfer accuracy [Key]
UTC time dissemination accuracy [Key]
Signal in Space Continuity [Optional]
Broadcast Polar Motion [Optional]
GNSS/RNSS Time Offset [Optional]
UT1-UTC Offset [Optional]
Carrier Phase Coherency [Optional]
Verifying the standards
The inevitable question one asks after considering these published performance standards is, are the service providers meeting their commitments? For several of the systems, the answer is generally yes, although insufficient data from monitoring and assessment services for each of the GNSS/RNSS is available to be conclusive. Both GPS and Galileo publish regular reports assessing performance against their standards. As early as 1993, the United States Federal Aviation Agency (FAA) began to publish quarterly performance analysis reports which scored the GPS performance against the existing standards (FAATC). In 2013, the U.S. Air Force began issuing public reports by the University of Texas Applied Research Laboratories that analyze GPS Standard Positioning Service performance (ARL:UT, Analysis of GPS SPS Performance ). In 2017, the European GNSS Service Centre (GSC) began publishing a quarterly assessment of Galileo initial operational service (GSC, Quarterly Performance Report).
By and large, the performance of these systems has been stellar, with quarter after quarter of full compliance with the published performance standards. In fact, when looking at the data, we see an interesting characteristic of the performance standards, namely that the actual performance of the system far exceeds the levels set by the standards. This “looseness” in the performance standards is so extreme that in several recent major service failures, the performance standards levels were not even violated, despite the fact that numerous users were severely affected. Following are two such occurrences:
• January 2016. Fifteen GPS satellites broadcasted erroneous values of the UTC offset correction term in their navigation message for a period of over 12 hours. As a result, thousands of digital radios and communication networks around the world went off line. The satellite operators were made aware of the problem, and they restored the signals to normal operation. In its report for the period, the University of Texas Applied Research Laboratories said, “Several GPS satellites broadcast flawed UTC offset information on 25-26 January 2016. The flawed data included tot values that did not satisfy the allowable range criteria. Therefore these data were not considered in this process. No values exceeding the NTE [not-to-exceed] threshold were found in 2016.” (ARL:UT, Analysis of GPS SPS Performance for 2016, 2016)
• July 2019. All satellites in the Galileo constellation were disabled from service for six days due to a technical incident related to the Galileo ground infrastructure. The satellite operators eventually were able to correct the problem and bring the service back online for its users. In its report for this period, the GSC reported only on a violation of UTC availability, but no violation of positioning service availability. The GSC said, “During this quarterly reporting period, the measured Galileo Initial Open Service performance figures exceed the Minimum Performance Level (MPL) targets specified in the [OS-SDD], with the exception of the UTC availability MPLs in July.” (GSC, Galileo Initial Services Open Service Quarterly Performance Report, 2019Q3)
What can we learn from the reports regarding these incidents? One is that in some cases the performance standards lack sufficient detail to allow flawed data to be broadcast without violating the standards. Another is that it appears that since many of the performance standards are statistical in nature, accumulated over many days, that significant service disruptions are obscured by the statistics. Too many good performance days can hide the terrible performance on a few bad days.
These incidents aside, one could conclude that these performance standards are in general not set as tightly as they could be. The GPS performance standard commits to 95% range error over all ages of data of 7.8 meters, yet operates at around 1 meter. Galileo’s service definition document commits to 95% range error over all ages of data of 7 meters, yet operates at 20-35 cm. Not all performance standards set such a high level for range accuracy, however. BDS sets its performance level for range error at 1m RMS (roughly 2m 95%) and QZSS at 2.6 meter 95%. GNSS services may wish to revisit their performance standards and set levels that provide sufficient allowance in random fluctuations, but do not hide or obscure serious performance events.
GNSS service providers are to be commended for providing performance standards, and verifying they are being met. With multiple constellations, there is an opportunity to learn from one another and improve the service commitments.
Multiple GNSS systems and services are being implemented to meet national goals. While offering largely the same service of positioning and time, these GNSS systems also provide independence and redundancy, which is invaluable to PNT users. They also provide a degree of competition, which spurs providers to improve their own services. In the next few years, I believe that all GNSS and RNSS will have documented performance standard commitments, and to a great extent, these performance standards will conform to the guidelines set by the International Committee on GNSS.
I also expect that every service provider will begin to assess and publish its performance against these standards. My hope is that as this information is shared, there will be motivation to tighten up the performance standards, so that no single GNSS falls behind the others in providing reliable and accurate service to their users. This, of course, will be to the benefit of those who rely daily on these multiple GNSS/RNSS services.
Disclaimer and acknowledgements
This article solely reflects the views of the author. The author gratefully thanks Brent Renfro for his help in proofreading the manuscript and providing guidance on clarifying the technical wording.
ARL:UT. (2016). Analysis of GPS SPS Performance for 2016. www.gps. gov/systems/gps/performance/2016- GPS-SPS-performance-analysis.pdf.
ARL:UT. (n.d.). Analysis of GPS SPS Performance. www.gps.gov/ systems/gps/performance/.
BDSOSS2. (2018). BeiDou Navigation Satellite System Open Service Performance Standard, (Version 2.0), December 2018.
FAATC, W. H. (n.d.). GPS SPS Performance Analysis Report. www. nstb.tc.faa.gov/DisplayArchive.htm.
GalOSSDD1. (2019). European GNSS (Galileo) Open Service Service Definition Document, May 2019.
GLOOSSP1. (2019).Стандарт эксплуатационных характеристик открытого сервиса (СТЭХОС), Global Navigation Satellite System GLONASS Performance Standard for Open Service, Revision 2.2, 2019.
GPSPERF2018. (2018). An Analysis of Global Positioning System (GPS) Standard Positioning Service (SPS) Performance for 2018.
GPSPS1. (1993). Global Positioning System Standard Positioning Service Signal Specification 1st edition. www.gps.gov/technical/ps/1993- SPS-signal-specification.pdf.
GPSPS2. (1995). Global Positioning System Standard Positioning Service Signal Specification 2nd edition, June 2, 1995. www.gps.gov/technical/ ps/1995-SPS-signal-specification.pdf.
GPSPS3. (2001). Global Positioning System Standard Positioning Service Performance Standard, October 2001. www.gps.gov/technical/ps/2001- SPS-performance-standard.pdf.
GPSPS4. (2008). Global Positioning System Standard Positioning Service Performance Standard, September 2008. www.gps.gov/technical/ps/2008- SPS-performance-standard.pdf.
GSC. (2019Q3). Galileo Initial Services Open Service Quarterly Performance Report. www.gsceuropa. eu/sites/default/files/sites/ all/files/Galileo-IS-OS-Quarterly- Performance_Report-Q3-2019.pdf.
GSC. (n.d.). Quarterly Performance Report. www.gsc-europa.eu/ electronic-library/galileoservice- performance-reports.
ICD-GPS-200V1. (1983). “Navstar GPS Space Segment / Navigation User Interfaces”, ICD-GPS-200.
ICGPSG. (2018). Guidelines for Developing Global and Regional Navigation Satellite Systems Performance Standards (Version 1.0). www.unoosa.org/documents/ pdf/icg/PS/Performance_Standards_ Guidelines_V1.0.pdf.
ICGWGA. (2011). Report of the Working Group A: Compatibility and Interoperability (ICG/WGA/2011). www.unoosa.org/pdf/icg/2011/ icg-6/wgA/ICG-6_WGA.pdf.
InsideGNSS. (2008). FAA Publishes WAAS Performance Standards. InsideGNSS Nov 10, 2008.
QZPS1. (2018). Quasi-Zenith Satellite System Performance Standard, November 5, 2018.