Revision of RAIM Implementation for Maritime

Abstract

The field of GNSS integrity comprises a plethora of integrity related algorithms, each featured with different characteristics. These algorithms use a priori information of some parameters in order to provide integrity information to the user. One type of these algorithms, RAIM, is extensively used in the maritime domain in order to provide an indication if the position calculated is likely to be outside of the requirements of maritime performance standards, as it is required by IMO regulation. The term “Classical RAIM” refers to fully autonomous algorithms which can implement only one, a combination or all the following capabilities: Fault Detection, Fault Exclusion and Protection Level computation. In order to provide this indication, a GNSS receiver shall incorporate RAIM using at least Fault Detection (FD) algorithms, or similar means to determine if accuracy is within the performance standards and provide an integrity indication. In order to harmonise and ensure the correct RAIM algorithms implementation, its usage is standardised by the standard IEC 61108, in particular 1 and 3 series for GPS and Galileo constellations. The current RAIM definition in standards does not constrain the specific RAIM algorithm implementation deliberately in order to let the manufactures implement the most suitable one for their applications. In addition, in order to prove that a receiver is compliant with the requirements specified in this standard, two simple tests are proposed. The first one intends to test the performance of the receiver under safe and caution states, and the other one tests the unsafe state. However, as it is recognised by IALA in the last workshop on the future of marine radiobeacons, “current implementations of Receiver Autonomous Integrity Monitoring (RAIM) are considered inadequate due to the current lack of maritime specific RAIM algorithms.” It has been observed some limitation of current RAIM implementations beyond a specific algorithm for maritime. First of all, the relaxed requirements, in terms of high probabilities of false alarms and of not detecting an error condition, might be justified due to the limitations expected from some implementations of RAIM algorithms. In addition, requirements for GPS and Galileo RAIM differs significantly. Secondly, they are based on some hypothesis as single-failure assumption, a priory estimation of error values components (too conservative in order to provide the integrity required for any user under any condition) or the no existence of biases in faultfree measurements that may not ensure the integrity required if these hypothesis are not met in maritime domain. Finally, considering the simplicity of the aforementioned tests, which do not provides a set of parameters for testing or even assess the probabilities of miss-detection or false alarm, there is no assurance that the RAIM algorithm implemented in the receiver correctly provides the integrity required. To ensure this, it would be necessary to improve the test method in order to characterize the value of the introduced error and add new tests to evaluate the complete set of requirements to the implemented integrity algorithm. The GSA (European GNSS Agency), under GSALOT3TRANS framework contract, analysed the implementation of EGNSS in the area of transport applications. In particular, the Specific Contract 7 (SC7) MarTech it is being assessed the potential use of SBAS combined with RAIM in maritime and inland waterways (IWW) domains. The industrial partners of this specific contract are VVA Brussels (prime) and GMV. Although is not contemplated in any of the aforementioned regulations, the use of SBAS on top of other GNSS technologies that use RAIM may imply some important advantages. The first advantage and the most obvious of the SBAS corrections usage on top of RAIM is the increased performances in terms of accuracy, availability and continuity. The use of SBAS integrity information may also overcome the limitation of some RAIM algorithms that are only able to detect one faulty satellite. SBAS integrity messages also contains valuable information about the estimation of error components that may be used by the RAIM algorithm on real time with enough level of integrity to improve the performance. The aim of this paper is to provide a summary of the activities developed so far as part of GSALOT3TRANS SC7 regarding the current RAIM limitations in the maritime domain and the potential use of SBAS in combination with it to overcome some of these limitations.