The Error Analysis Of The Virtual Reference Station Algorithm

The Virtual Reference Station (VRS) concept is one of the most latest promising concept, in a broad sense of using multiple reference stations ,which is applicable for centimeter-level accurate differential GPS positioning in real time environment, To create a VRS i.e. a reference station near by the rover receiver, the task concentrates on simulating a series of reference data, including both the accurate position and the GPS carrier phase observations, based on the benefit of using all of the observed information. Compared with the conventional single station applications, e.g. RTK, Multi-Base methods or VRS technique possesses many virtues: wider action area, more accurate, more dependable positioning, and so on. So VRS technique is gradually becoming a focus in research of the accurate position by GPS. It is investigated with the emphasis on the spatial correlated error and the methods of creating correction for VRS in this paper.The research and main achievements of this paper are as follows:1. Firstly, the influence factors and characteristics, which satellite orbit errors act on baseline, are analyzed in theory. Then simulated experiments are done by using different satellite orbits, such as Broadcast orbits, IGS Final orbits, IGS Ultra-Rapid orbits. After discussed how the factors like satellite elevation and distance length influence on positioning, a conclusion that there is a better linear correlation between observations of X and Z axis vector of baselines and the length of them is drawn. It is advised that IGS Ultra-Rapid orbits or CODE predicted orbits can be made use of when someone deal with the high precise and real-time positioning by VRS.2. It is given that the relations between satellites elevation and troposphere delay of observations. After calculated and analyzed, it is also shown that there is a strong linear relation between baselines of VRS and single-difference troposphere delays, the ionosphere delays also have. In addition, the factors, such as temperature, air pressure, water vapor, difference of altitude and model, can also influence on single-difference troposphere delay.3. The methods which can create corrections terms of moving stations are developed by introducing the factor of weight, which is on the condition of different influences on creating corrections of spatial error in moving stations made by references stations in network. It is shown that the positioning precision can be improved by introduced reasonable the factor of weight.4. In addition, the optimization triangle interpolation algorithm which is used in mathematics usually is introduced to creating correction terms in VRS for the first time in this paper.