The Application Of Triple-frequency Data Combinations In Cycle Slip Detection And Repairing

Future global navigation satellite system (GNSS), such as the US modernized globalpositioning system(GPS),the Russian GLONASS system, the China COMPASS system and theEuropean Galileo system, will all operate with three or more frequencies, owing tomulti-frequency data, more data combinations with superior property are available for cycle slipdetection. In this paper, on the basis of researching single and dual frequency cycle slip detectionand repair methods, based on triple frequency data can form more combinations with excellentproperties，such as long wavelength、ionosphere free or low ionosphere and low noise, putsforward two new methods for the modernized GNSS cycle slip detection: By choosing twogroups of geometry-free phase combinations (-1,-1,2) and (-1,4,-3), we can detect all cycle slipsexcept sensitive cycle slips in200cycle, but the combined cycle slips value can not be fixed toan integer, the cycle slips on the basis of carrier can not be separated also，with a third-orderpolynomial fitting to single difference in frequency ofL1、L2andL5, these four sets of equationsin accordance with the principle of indirect adjustment, could effectively detect cycle slips smallto one cycle at every frequency, this method is suitable for high sampling rates cycle slipdetection; A method for optimal selection of the triple-frequency combination test quantities isinvestigated based on the theory of phase minus pseudorange combination for cycle slipdetection, three linear independent carrier phase combinations are selected in the case ofdifferent pseudorange measurement noises, the combination test quantities can accurately detectcycle slip small to one cycle at every frequency.Detecting the cycle slip of original observation data of L5NetR8Test station in IGSwebsite,found no cycle slip, the sampling interval is15s, adding simulated big cycle slip、smallcycle slip and special cycle slip or insensitive cycle, both the two methods detected the cycleslips correctly. Only up to two sets of linearly independent combinations are formed bygeometry-free phase combination, due to the error of polynomial fitting to single difference infrequency of15s sampling rates is up to0.6cycle, sometimes may cause rounding errors. Thetriple phase minus pseudorange combination with the cycle slip valuation standard deviation ofabout0.2cycle, even though under condition of3meters standard deviation of pseudo-range, thecombined cycle slip can still be rounded success probability of95.5%.