Preselection Method of Ultrafast Clock Error Prediction Model for Satellite Clocks

Abstract

The invention relates to a method for preselecting a clock error prediction model of a satellite clock, and belongs to the technical field of clock error prediction preselection of satellite clocks. Technical solution: Use the fast clock error products provided by IGS to extract the change information of the relative frequency deviation across the sky, fit the change information of the relative frequency deviation across the sky of the fast clock difference data, and obtain the forecast of the relative frequency deviation of the fast clock difference data value; the cross-sky relative frequency deviation of the first-order polynomial forecast value and the cross-sky relative frequency deviation of the quadratic polynomial forecast value are respectively compared with the forecast value of the cross-sky relative frequency deviation of the fast clock error data, and the absolute value of the two differences is taken value, select the model with smaller absolute value as the prediction model of ultrafast clock error. The invention can pre-select the prediction model of the ultra-fast clock error, and the RMS of the forecast data of the selected prediction model is relatively the smallest, and is not limited by the delay time of about 14 days of the final clock error product, and provides the relatively smallest RMS in time forecast model.

Description

technical field [0001] The invention relates to a method for pre-selection of a clock error prediction model of a satellite clock, which utilizes the relative frequency deviation variation characteristics of the clock error data of the satellite clock to realize the pre-selection of a clock error prediction model, and gets rid of the requirement of the final clock error data with the highest accuracy. The time limit of about 14 days delay makes the root mean square error (RMS) between the forecast data of the pre-selected forecast model and the final clock difference data of the corresponding day to be the smallest, which makes up for the clock difference forecast technology in the field of forecast model pre-selection technology. The blank of the satellite clock belongs to the technical field of pre-selection of clock error prediction of satellite clocks. Background technique [0002] The clock error prediction of satellite clocks in the Global Navigation Satellite System (...

AI Technical Summary

Problems solved by technology

The problem is that the final clock difference product of IGS will be released after a delay of about 14 days. That is to say, after the ultra-fast observation clock difference data is forecasted, the forecast accuracy will not be known in time, let alone which one is known in advance. The RMS of the polynomial model is relatively small

Images