1. Absolute Receiver Autonomous Integrity Monitoring (ARAIM)
Todd Walter
Stanford University

2. Introduction
GPS is an important component of today’s aviation navigation infrastructure
Its role will continue to increase over the coming years
Future GNSS constellations will also become important contributors
However, their incorporation must be done with great care as the integrity requirements for aircraft guidance are very stringent
Less than 10-7 probability of misleading information
International standards define different types of GNSS augmentations to achieve this level of integrity

3. Integrity Monitoring
Satellite-based and ground-based augmentation systems provide independent monitoring of the GPS signals through calibrated ground monitors
Requires ground monitoring network communication channel to aircraft
Receiver Autonomous Integrity Monitoring (RAIM) compares redundant satellite range measurements against each other to identify and eliminate significant faults
Requires a greater number of ranging measurements than SBAS or GBAS

4. ARAIM Protection Level
Compass
GLONASS
Galileo
VPL
GPS
VPL
VPL
VPL

5. RAIM vs. ARAIM LNAV requirements are much less stringent than LPV
Alert limit measure in nautical miles
Only real threat is a large clock error
For LPV MI is hazardous (vs. major)
Alert limit in tens of meters
Many sources of potentially significant errors
Two or more smaller errors may combine to cause a large enough error
ARAIM needed to more carefully account for all threats

6. Interoperability of Integrity
Interoperability should be a goal not only for GNSS signals, but also for integrity provision
Augmentation systems already internationally coordinated
Open service signals should target performance comparable to or better than GPS L1 signals today
Different service providers may make different design choices and different assurances
However, it is important to establish a common understanding of how RAIM depends on GNSS performance and how signals from different services could be combined to improve RAIM
Cooperation and transparency are essential

7. Benefits of Multi-Constellation RAIM
Combining signals from multiple constellations can provide significantly greater availability and higher performance levels than can be achieved individually
Support for vertically guided approaches
Potential to provide a safety of life service without requiring the GNSS service provider to certify each system to 10-7 integrity levels
Creates a truly international solution
All service providers contribute
Not dependent on any single entity
Coverage is global and seamless

8. Service Commitment
Each service provider should provide documentation of their service commitment
Encourage usage by other states
Allows planning of combined service level
Supports development of interface specifications and user algorithms
Commitment should include details on:
Accuracy, continuity, availability, fault modes, broadcast parameters, and other operating characteristics
Assurances should be provided for minimum commitments

9. Specification of Faults
Perform a fault modes and effects analysis
Understand and make transparent potential faults and their effects
Assure low fault rates
Of order 10-5/SV/Hour
Assure low probability of simultaneous or common mode faults
Ideally below 10-8/Constellation/Hour
Assure a short time to alert
Not longer than 6 hours
Maintain independence from other service providers

10. Monitoring and Assurance
Methods for monitoring conformity of signal properties relative to provided assurances should be agreed upon mutually by service providers and approval authorities
Require clear unambiguous evaluations of assurances
May be made by any potential approval authority
Desirable to have a means to resolve potential observations of non-conformity
Long-term monitoring by each sovereign state is an important component of establishing reliability
Each constellation still cross-checked by others in user avionics

11. Interface Specification
Each system may broadcast different parameters or provide different levels of assurance
However, a common understanding of how each parameter is used must be reached
The parameters must be combined into a single upper bound for the joint position estimate
The upper bound must be safe regardless of which combinations of satellites are used
Also able to account for potentially different properties
Requires a more advanced form of RAIM than is used currently for LNAV
Good candidates already exist

12. Summary
RAIM allows for worldwide aviation navigation without requiring additional ground infrastructure
Additional GNSS constellations can significantly improve performance and availability
At a minimum, new GNSS constellations should assure that their open service signals support existing LNAV RAIM
Should work together to specify a means to achieve multi-constellation RAIM for vertical guidance
International cooperation and coordination will be essential to achieving this goal

13. Specification of Accuracy
The dominant error sources should be understood and characterized
Satellite clock and ephemeris within constellation tied to clear, stable, global, reference frames
Code and carrier signals coherently derived from a common source
Well designed signals to reduce multipath, ionosphere, and distortion effects
International coordination already well-established
Document how the expected performance level is indicated to the user
Should broadcast expected accuracy for a fault-free ranging source

14. Specification of Availability & Continuity
Description of constellation geometry
Number of satellites, planes, spacing, etc.
Assure minimum levels of operating satellites
e.g probability of at least 20 primary slots occupied by satellites broadcasting valid signals
Assure minimum levels of continuity
e.g. less than probability of unscheduled interruption or fault of previously healthy signal
Lesser minimums support multi-const. RAIM even if they cannot support stand-alone

15. Fault Tree and Probability of Hazardously Misleading Information (PHMI)
Courtesy:
Juan Blanch
Any mode causes HMI
PHMIk
PHMI0
PHMI1
No failures/ rare normal create HMI
failure of sat 1 causes HMI …
failure of sat k causes HMI …
Mode prior probability = ~1
Mode prior prob-ability = ~1e-4
Mode prior prob-ability = ~1e-4
For each branch, a monitor mitigates the probability of HMI given the failure
In ARAIM, the monitors are formed by comparing subset solutions