Research On Fusion Algorithm Of Cycle Slip Detection And Correction For GNSS

With the rapid development of GNSS technology, it will be an era of multiple system fusion of global satellite navigation system in the future. The fusion and processing of multimode GNSS data will be one of the main development trends in the field of navigation and positioning. Cycle slip detection and the whole cycle ambiguity solution is the core issue of GNSS positioning in GNSS data processing. This paper mainly studies the multiple system cycle slip detecting and repairing fusion algorithm. Taking GNSS observations as the research object, the data quality of different satellite navigation systems is analyzed in details. Due to the different phase noise of different satellite systems, when the multimode GNSS data was cycle slip detected, it is difficult to set up a unified threshold condition. We proposed an innovative adaptive dynamic threshold method. GNSS cycle slip detection model was constructed through the ionosphere residual method and the improved double difference method. The main research contents include the following aspects:(1). The data quality of different satellite navigation systems was analyzed in detail. The multipath effect of GPS satellites, BDS satellites and satellites in the orbit of Big Dipper, and signal-to-noise ratio are analyzed in detail. For all the satellites, the lower elevation angle is, the relative higher multipath effect value is. As the elevation angle increases, multipath effect reduced obviously. Both frequency point B1 and B2, multipath effect presents:GEO< IGSO< MEO.(2). In addition to cycle slip and gross error, GNSS observations also contain the receiver clock jump. Clock jump is an important factor of disturbing gross error and cycle slip detection, it can lead to the failure of existing part of the cycle slip detection method. This paper proposes a method of clock jump detection on receiver clocks. The O-C values of double-frequency phase observations were adopted to conduct clock jump detection through the ionosphere residual method.(3). In the actual process of cycle slip detection, because of the influence of ionosphere activity and the sampling rate, the reliability of the threshold which depending on the prior value is decreased as the judgment of whether the cycle slip occurs. The phase noise of different satellite systems is different. It is difficult to set up a unified threshold condition when detecting multimode GNSS data cycle slip. We proposed an innovative adaptive dynamic threshold method according to the principle of polynomial fitting and made a detailed exploration with the ionosphere residual method as an example. We applied the method of threshold value structure to MW combination method and improved double difference no ionosphere combination method to conduct cycle slip detection. Experiment suggested that the dynamic threshold method can accurately detect all cycle slip, and accurately identify outliers.(4). During the process of processing the observed data, the principle of detection first and repair second was adopted. We construct a detection model of GNSS cycle slip detection by methods of second geometry free and double difference LC combination. According to the principle of polynomial fitting, the floating point solution of cycle slip with deviation marker was estimated by least square method of low-order polynomials. Finally, the best combination of setting integer cycle was searched through the search space chi-square test and it is the size of a cycle slip on the dual-frequency carrier phase. Experiment suggested that the detection model of GNSS cycle slip can accurately detect the full cycle and accurately identify outliers. The repair model of GNSS cycle slip can correct repair single epoch combined cycle slip, but the repair result of the continuous cycle slip is not correct.