Research On Key Technologies Of GPS/BDS High Precise Relative Positioning

The BDS navigation satellite system has started to provide the PNT(Positioning,Navigation and Timing)services for the Asia-pacific region at the end of 2012.At the same time,the BDS three generation satellite system is still in the stage of network construction currently,and is expected to provide global PNT services in 2020.However,in this current stage,there are very few satellites which can be observed for single BDS system and the coverage area is limited.In addition,the performance of the hardware of the new generation system is required to further verification.Therefore,applying the GPS and BDS system simultaneously,making their advantages complementary and information exchange,not only can increase the number of satellite observations and improve the positioning accuracy and efficiency,but also can avoid the risk of relying on single GPS system.Specifically,the main research contents and contributions for this dissertation includes the following aspects:Considering the existing TurboEdit cycle slip detection and repair algorithm MW combination cannot detect the small cycle slip and continuous cycle slip accurately,and the GF detection combination produce the false cycle slip frequently,which leads to the cycle slip is difficult to be repaired.The cycle slip detection and repair SET(Satellite Elevation TurboEdit)method based on un-difference dual-frequency observations have been proposed.The key idea of this new SET method mainly includes two parts:1)using the adaptive sliding window for MW combination detection model to establish the connection between the length of the sliding window of the MW combination and the satellite elevation of the current epoch,which can make the multipath error and noise error directly related to the length of the sliding window.Thus,by adjusting the length of the window adaptive,the small cycle slip and special combined cycle can be detected accurately.2)using difference by epoch method instead of the difference by pseudo-distance for GF combination and also introducing the elevation weighting coefficient for the detection threshold,which can eliminate the false cycle slip and detect the cycle slip effectively.The experimental results shown that the proposed SET method can detect the small cycle slip,the big cycle slip,the special cycle slip and the continuous cycle slip accurately,and repair the cycle slip effectively.We compared the three commonly used ambiguity decorrelation algorithms of GNSS by the time efficiency analysis,the results shown that the the LLL-IGS algorithm performs unstable and the efficiency is much lower than the others,especially for the real GNSS data.In order to solve this situation,a modified LLL-MIGS decorrelation algorithm is proposed by improving the sorting method to reduce correlation of the co-variance matrix and removing the error of orthogonalization by adjusting the order of the production.In addition,the time efficiency is also introduced as a new assessment criterion to measure the performance of the decorrelation algorithm directly,which includes the decorrelation time efficiency and searching time efficiency.At last,simulation date and real GNSS observations which including short baseline,network-based medium and long baselines have been used to compare the LLL-MIGS with LLL-IGS and also to analyze them in depth.The results of the experiments shown that the LLL-MIGS method performs much better than LLL-IGS method in decreasing condition number and reducing time consumption which includes the decorrelation time consumption and searching time consumption.Moreover,both of them indicate that the modified LLL-MIGS algorithm is more stable than the traditional LLL-IGS method.Since the existing carrier multipath correction algorithms cannot extract the multipath correction model accurately,the repetition of the satellite orbit affected by satellite maneuver seriously,the elevation and azimuth angle model dependently the resolution of the correction model and the sampling rate of the GNSS data,a new single-difference filtering multipath correction method based on adaptive wavelet transform denoising and double reference transfer strategy is proposed.The key idea of the proposed method is the use of both the adaptive thresholding wavelet denoising to extract an accurate multipath correction model from reference Day and the double reference shift strategy to mitigate multipath for subsequent Day2 more accurately and efficiently.The new method includes three parts,the double difference transfer to single difference model,the discrete wavelet transform for denoising model and the orbital repetition of the satellite calculation model.Two groups of ten days GNSS observation dataset have been used to analyze the orbital repetition period of the satellites,the results of the experiments shown that the mean shift time of each satellite is different from each other and the difference between maximum and minimum is about ten seconds.Furthermore,for the same satellite,although the mean shift time is different for every day,the changes are very small.In addition,two short baselines have been used to assess the performance of the ATDR method,the results of the experiments shown that the 3D positioning precision is improved by about 37.73%for the single epoch mode with 30s data sampling rate and about 31.22%for the continuous mode with 1s high sampling rate compared with the original results.Considering the problem of the pseudorange multipath deviation of the BDS satellite-induced,the observation collected by MGEX data center have been used to analyze this multipath deviation.The results shown that the performance of the satellite-induced multipath bias of the IGSO and MEO satellites is different from each other for individual,and the bias is closely related to the elevation angle of the satellites.However,the GEO satellites are not related to the elevation tightly.In addition,the orbital repetition of the three type of BDS satellites have been analyzed.Therefore,considering this situation,the modified correction DORC method have been proposed.The new method establish the correction model for each IGSO and MEO satellite individual for correcting the satellite-induced multipath bias,and use the sidereal filtering method for correcting the GEO satellite-induced bias.The experiment results shown that the proposed DORC method can reduce the satellite-induced bias effectively,for the IGSO and MEO satellites individual correction model,the improvement rate of the DORC correction model can reach by 21.05%,and for the GEO satellites sidereal filtering method,the improvement rate of the DORC correction model can reach by 34.15%.In addition,for the HMW combination,the mean improvement rate of the HMW combination can also be improved by 29.71%.