Research On Stochastical Model For Single Point Positioning Of GPS&BDS Conbined System

At present,using different global navigation satellite system(GNSS)for combined system positioning has been the development tends of the global navigation and positioning technology fields Combined satellite system positioning can greatly improve the situation of insufficient satellite observations in a single navigation system,thus enhancing the spatial geometric distribution structure of satellite constellation,which is superior to single system positioning in terms of navigation and positioning stability,reliability and usability.GPS in today's GNSS application field still occupy an dominant position,widely used in the world.But GPS is established and controlled by the United States,and has the ability to selectively shut down GPS service,so other countries' navigation services accuracy or dependency cannot be guaranteed.Therefore,on the principle of peaceful development and compatibility,China has also developed a completely autonomous Beidou satellite navigation system(BDS),and the current BDS has been able to provide the Asia-Pacific region navigation and positioning services which can meet the daily needs of the application accuracy.Research on GPS and BDS combined system positioning can not only improve the availability of navigation positioning,but also is conducive to the development and improvement of China's Beidou system.The GNSS positioning adjustment model includes function model and stochastic model,in which stochastic model has a great influence on the final positioning result,it is necessary to determine a reasonable stochastic model-the weight matrix.GPS\BDS combined systems satellite observations are from two different systems,thus the different system has different observation accuracy,and because the different satellite signal transmission path will cause different measurement error,so the different satellite observation's quality is also different even from the same navigation system.Therefore,we can determine the GPS\BDS positioning weight matrix through those two aspects.The basic principle of weight determination is based on the contribution of each observation to the positioning accuracy,so give a larger weight to the more useful observations.In the GPS\BDS combined system,the contribution of each observation to the final solution' accuracy is determined by its own noise.The common fixed weight models includes the satellite elevation model and SNR model.On the determination of the weight between different systems,the more advanced method is Helmert variance component estimation.For the GPS/BDS combined positioning,we must establish a suitable stochastic model to determine a weight matrix of different systems and different satellite observations.In this paper,based on the pseudo-range single point positioning method to study the combination of positioning,evaluating the navigation and positioning capabilities between the single-system and combined system,and the current commonly used GNSS fixed model for comparative analysis,proposed a new stochastic model of GPS\BDS combined positioning with a method using entropy weight to integrate the satellite elevation model and the SNR model,and combine with the Helmert's posterior variance component estimation method.The experimental results show that the above stochastic model can effectively improve the GNSS positioning accuracy,and GPS\BDS combined system still has the positioning accuracy within 10m when the single system cannot be positioned due to the lack of satellite observations.