关键词: 网络RTK, 整周模糊度固定, 随机模型, 电离层误差, 功率谱密度

Abstract:

The integer ambiguity resolution of the reference station network is the basis of high precision positioning of network RTK. However, with the increase of the range of the reference stations, the spatial residual of atmospheric error makes it difficult to resolution reference station, especially the ionospheric delay error with complex temporal and spatial variation, which seriously affects the resolution performance of the reference station network. The atmospheric parameters are estimated by random walk when the ambiguity resolution of the reference station. Based on the analysis of the ionospheric power spectral density (IPSD) on the ambiguity resolution performance of the reference station, this paper studies the time-varying characteristics of ionospheric observations with different differential intervals. By observing the different trends of noise and ionosphere with differential time intervals, the ionospheric observation noise is weakened to determine the IPSD, and the stochastic model of ionospheric parameters in the ambiguity estimation is optimized, so as to improve the fixed efficiency of the long-distance reference station network, instead of using empirical values or empirical models that do not consider atmospheric variation. The experimental results show that the IPSD estimated in real-time by the 1 s sampling interval data can optimize float solution accuracy of the integer ambiguity of the reference station, and can also reduce the search space of the integer ambiguity. Compared with empirical ionospheric power spectral density, the method proposed in this paper can improve the convergence time by 21% in five reference station networks with a baseline length of more than 100 km, and the success rate of ambiguity resolution is also improved accordingly.

Key words: network RTK, integer ambiguity resolution, stochastic model, ionospheric error, power spectral density