A Holistic Approach to Carrier-Phase Receiver Autonomous Integrity Monitoring (CRAIM)

Carl Milner, Shaojun Feng, Altti Jokinen and Washington Ochieng, Chris Hide, Chris Hill and Terry Moore

Abstract:	Pseudorange-based integrity monitoring such as Receiver Autonomous Integrity Monitoring (P-RAIM) and its variations, has been studied extensively over recent decades. This has primarily been driven by the safety critical nature of aviation and the important link to the integrity of the positioning and navigation solution required. However, for higher accuracy applications, the more precise carrier phase measurements are used. The applications include positioning both in static and dynamic modes, with the latter employing Real Time Kinematic (RTK) positioning techniques. Furthermore, for mission critical applications such as airport surface movement in which safety is critical, user receiver level integrity monitoring akin to P-RAIM for carrier phase data is required. The state-of-the-art in user receiver level integrity monitoring with carrier phase measurements is formed of two distinct steps; ambiguity validation and a traditional RAIM consistency checking step which assumes fixed ambiguities. This paper proposes a holistic Carrier phase RAIM (C-RAIM) technique that integrates the two processes. The key to accurate and reliable carrier-phase positioning is reliable determination of integer phase ambiguities. A number of methods are currently used, the most popular being the Least-squares AMBiguity Decorrelation Adjustment or LAMBDA, as well as the simpler but sub-optimal integer rounding and sequential integer rounding or bootstrapping. This is followed by a process to confirm the quality of the resulting ambiguities, referred to as ambiguity validation. The main approach to ambiguity validation uses statistical tests such as the ratio test, F-test, W-ratio and ?2-test which are not justified in theory. However, recent studies have looked to characterise the distribution of the test statistic or incorrect ambiguity vector more accurately in order to provide an estimate of confidence in the ambiguity resolution step. These include; the Integer Aperture (IA) method with Monte Carlo sampling, the doubly-non-central F distribution and Bayesian statistical techniques. The IA approach is computationally demanding and therefore, is unsuitable for real-time applications. The doubly non-central F distribution method determines the confidence level as a function of the geometry and ambiguity residuals from the LAMBDA method. The Bayesian approach determines the probability of the ambiguities being observed given a set of candidate ambiguities whose size may be varied to requirements. The existing carrier phase based user level integrity monitoring techniques above have a number of weaknesses associated with reliability and robustness. These include the assumption of correctly fixed ambiguities on the basis of ambiguity validation tests, a lack of consideration for cycle slips before and after fixing, noise correlation as a result of the double differencing operator and a separation of the ambiguity validation and consistency check components. This paper presents a CRAIM method which addresses these issues by firstly estimating the ambiguities within a Kalman filter. The double differenced pseudorange, multi-frequency linear combinations and carrier phase measurements are used as observables in the Kalman filter. A multiple hypotheses approach is then used to derive the current integrity performance in terms of protection levels. The ambiguity validation methods which provide estimates of the probability of correct ambiguity resolution are used to assign the integrity risk between multiple hypotheses. This approach ensures that the two integrity monitoring steps of ambiguity validation and consistency check are combined into a process which outputs a single performance metric. This is consistent with the notion of a holistic integrity monitoring concept. The assumption of correctly fixed ambiguities is thus dropped as in previous approaches and the impact of undetected cycle slips is accounted for. The integrity allocation between the hypotheses is then optimised in the protection level domain to improve the availability and continuity of potential applications. GPS and Galileo data generated by the Spirent Simulator is used to test the performance of the algorithms injected with a variety of failure models. The paper shows how the integration of the two integrity monitoring techniques enable the optimisation of the integrity risk requirements thereby resulting in lower protection levels. This leads to improved availability and preservation of continuity, with the benefit that more stringent applications can be supported.
Published in:	Proceedings of the 24th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS 2011) September 20 - 23, 2011 Oregon Convention Center, Portland, Oregon Portland, OR
Pages:	2689 - 2695
Cite this article:	Milner, Carl, Feng, Shaojun, Jokinen, Altti, Ochieng, Washington, Hide, Chris, Hill, Chris, Moore, Terry, "A Holistic Approach to Carrier-Phase Receiver Autonomous Integrity Monitoring (CRAIM)," Proceedings of the 24th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS 2011), Portland, OR, September 2011, pp. 2689-2695.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In