New Approach for Integrity Risk Evaluation Based on Expected Acceptable Region for Carrier-Phase-Based Precise Relative Positioning

Abstract GNSS carrier-phase observation is widely adopted to achieve centimeter-level relative position accuracy in life-critical or mission-critical GNSS applications. Different from code-based relative positioning, carrier-phase-based precise relative position still suffers from insufficient integrity risk evaluation caused by its inherent risk sources such as cycle slips and ambiguity resolution. This paper proposes a novel integrity risk evaluation approach for relative positioning, especially considering cycle slips. To eliminate cycle slips under the condition of concurrency in multiple channels more efficiently, INS-aided information is adopted for cycle slip detection and elimination. To reduce the difficulty of deriving the post-test variance of observations and simplify the process of integrity risk evaluation, the concept of Expected Acceptable Region (EAR) is first proposed to decompose total integrity risk as two independent components, i.e., intra-EAR risk and extra-EAR risk. Different evaluation algorithms are adopted based on the characteristics of different regions. To verify the superiority of the proposed method, numerical experiments have been conducted to simulate the aircraft's precision approaching and landing, where the proposed algorithm is adopted to meet the strict integrity requirements for Sea-based JPALS. It is indicated that the proposed method can provide both conservative and accurate estimation of integrity risk with a small part of the prior probability of cycle slip (i.e., \(\pm 1\)cycle). Compared with carrier-phase-based relative positioning in the ARAIM frame, the proposed method can increase GNSS availability by about 20% with a significant decrease in computation time by about 76%.