系统分析了改进随机模型和改进函数模型两类GPS基线解算模型的优缺点,在此基础上,提出了一种基于序列平均的高精度GPS基线解算模型,即采用动态单历元技术进行静态基线解算,充分利用多路径效应的低频特性,采用小波变化理论,对坐标序列进行多路径效应的去除,提取低频残差项进行序列平均,得到基线向量解。同时,以动态坐标序列为依据,对出现粗差历元或者卫星进行处理,有效弥补了仅采用残差序列进行粗差判断的不足,提高了基线解算的精度和可靠性。实验表明,新模型可以更为有效地削弱多路径效应的影响,而且对于较短的观测时间尤为突出;结合坐标序列和残差序列,能更为有效地进行粗差的探测和去除,提高基线解的精度和可靠性。
Multipath and gross errors are the most important factors of high-precision and high-reliability GPS baseline solution, especially for short period and low sampling rate. Two main kinds of traditional GPS baseline solution models are introduced, on the basis of which, a new GPS baseline solution model, utilizing single epoch methodology to obtain baseline solution, is proposed. It is thus solved by reducing multipath errors of coordinate series taking full advantage of low sampling rate of multipath errors and adopting waveletmulti-scale method to extract and eliminate the multipath error in prior to averaging the re- sidual series to obtain the baseline solution. Meanwhile, the precision and reliability of base- line solutions are improved by remedying gross errors of epochs, based on dynamic coordi- nate and series residuals series. The results of tests confirm that multipath errors are elimi- nated to a great extent, utilizing the new model, especially for short period, and the preci- sion and reliability of baseline solutions are enhanced by detecting and reducing gross errors on the basis of combination of coordinate series and residuals series.