Algorithm Research On GNSS Receiver Autonomous Integrity Monitoring

With modernization of GPS and implement of Russia’s full constellation of GLONASS, the development of Galileo by the European Commission and European Space is continuing. The positioning accuracy will be significantly improved due to the future multi-constellation and multi-frequency. Achieving the high positioning accuracy, one pays more attention to the positioning quality. The quality control can be accomplished by GNSS user end such as Receiver Autonomous Integrity Monitoring (RAIM). RAIM refers to integrity monitoring of GNSS navigation signals performed by receiver independent of external reference systems, except of the navigation signal themselves. With lots of advantages such as no need of external facilities such as Ground-Based Augmentation Systems (GBAS) or Space-Based Augmentation Systems (SBAS) and concerning receiver around obstacles, RAIM draw attentions and significant efforts have been made to develop and analysis GNSS RAIM methods and algorithms over the past decades.While existing RAIM techniques, generally based on single bias will not be justifiable of the next generation GNSS due to multiple biases occurring with increased probability. An ideal RAIM can detect and identify biases with both of low false alarm rate and missing detecting rate, and meanwhile, RAIM provides user the HPL and VPL (Horizontal and Vertical Protection Level).This paper aims at improvement of RAIM algorithm for general GNSS user and focuses on evaluating RAIM performance based on Protection Level and Separability. Besides, the paper mainly seeks the algorithm for fault detection and identification for RAIM. The major contributions of this research are:a) Statistic characteristics of estimates were derived by supposing unperceived model error. The internal reliability and external reliability of Kalman filter were derived. Besides, the correlation coefficient for statistic test of Innovation is conducted.b) A new algorithm was proposed to calculate HPL and VPL for improving RAIM availability and Separability should be concerned for RAIM performance. Comprehensive analysis of GPS and GPS/GLONASS benefits with respect to RAIM was conducted. Some scenarios were performed to show how HPL, VPL and separability vary with satellite geometric, observation variance and probabilities of false alarm and missed dectection for single epoch, global snapshot and 24-hour temporal scenarios.c) Innovation sequence detecting bias method was presented to detect and indentify bias. Innovation sequence detection verified by comparing with conventional forward-backward method and correlation-test based forward-backward method. The performances of different FDI method were evaluated by correct detection rate, false alarm rate, correct identification, availability and positioning accuracy after exclusion. The performances analysis showed innovation sequence detecting can properly detect and identify not only single bias but also double biases even under only GPS.