Satellite clock error forecasting method

Abstract

A satellite clock error forecasting method comprises the following steps: step 1, acquiring an original clock error data sequence, and carrying out data preprocessing; step 2, performing primary ratio processing on the preprocessed clock error data sequence to obtain a new modeling data sequence; step 3, constructing a gray model in a discrete form according to the newly obtained level ratio data sequence; step 4, calculating parameters of the discrete grey model according to a least square method, performing level ratio forecasting according to the calculated discrete grey model, and fully mining grey information in the modeling clock error data sequence; and step 5, carrying out starting point deviation correction, and carrying out reduction processing according to the corrected model so as to realize satellite clock error high-precision prediction.

Description

technical field [0001] The invention relates to the technical field of satellite navigation and positioning, in particular to satellite clock error prediction technology. Background technique [0002] Satellite clock is one of the core loads of GNSS, its performance and the accuracy and reliability of satellite clock error prediction are the key factors affecting the positioning, navigation and timing service capabilities of the entire system. Without the support of the operation and control system for a long time, satellite autonomous navigation technology needs a segment of satellite ephemeris and clock error predicted by the ground as a priori information; in the process of real-time precise single point positioning, the clock error prediction result is required to participate in the calculation to achieve high Accurate positioning; time synchronization is also the basis of network operation and node coordination, and is an important guarantee for the realization of a lar...

AI Technical Summary

Problems solved by technology

[0004] Although the above-mentioned improved technologies have improved the prediction accuracy, they have not considered the gray information in the mining clock error data, and the stability of the improvement effect is poor.

There is a certain irrationality in the setting of the gray model’s essential principle. Its essence is to use an exponential curve to fit the cumulative generation sequence of the modeling data sequence, and then use the fitting function to predict the cumulative sequence. In fact, The one-time accumulation data processing method can make the accumulation sequence show a single-increase trend for a non-negative sequence, but it cannot guarantee the existence of an exponential change rate, and the change rate is not constant. Therefore, in the actual clock error forecast, the gray model sometimes appears relatively Large errors, or even completely lose the forecasting effect

This limits the fitting and forecasting accuracy of the gray model, and also contradicts the high-precision requirements of precise clock error forecasting and satellite navigation positioning

Images