Modeling the Ionospheric Delay based on Collocation for BDS Wide Area Augmentation System

Ionospheric delay is one of the most critical errors for various applications of GNSS (Global Navigation Satellite Systems, including American GPS, Chinese BDS, Europe Galileo and Russia GLONASS). Currently, the grid-based ionospheric map broadcasted by the Wide Area Augmentation System (e.g. American WAAS, European EGNOS, Indian GAGAN, and Japanese QZSS) or released by some organizations (e.g. IGS, International GNSS Services) via internet is the major source for obtaining precise ionospheric corrections. Considering the local characteristics of ionosphere over China and its surrounding area, a new approach for precisely modeling the ionospheric delay is developed for BDS WAAS. In this approach, the ionospheric delay is considered as the sum of the trend and stochastic terms. The trend term is represented by an adjusted spherical harmonic (ASH) function, and the stochastic term is described by a time-variant covariance function. The ASH-based ionospheric model and the covariance function are simultaneously estimated using real data in order to capture the subtle variations in local ionospheric delay. The real GPS data collected from about 30 reference stations in China were used to validate the proposed method for ionospheric modeling. The presented result shows that (1) the stochastic ionospheric delay can be depicted by the estimated covariance function and consequently improve the accuracy of grid-based ionospheric delay estimates; (2) the accuracy of ionospheric delay provided by the proposed approach is about 0.5m and 0.3m in the southern and northern areas of China with an average spacing of 600-1200km between permanent stations, respectively; (3) the convergence time of standard precise point positioning is shortened by about 20-30%, equivalently 8-15minutes, using the ionospheric delay correction from the proposed method.