Research On Rubidium Clock Taming Technology Based On Satellite Common View System

With the increasingly diversified application scenarios of mobile communication,new services such as high-precision positioning and industrial interconnection are demanding higher time synchronization accuracy.As an important part of the time-frequency system on board the global navigation satellite system(GNSS),rubidium clock's performance is the premise and foundation to ensure the accuracy of time-frequency transmission.However,the atomic clock is easily affected by environmental factors such as temperature,humidity,magnetic field,vibration and so on,which causes frequency drift and aging.Therefore,considering the continuous improvement of the coverage and working performance of navigation satellites,it has important theoretical value and broad development prospects to study the rubidium clock taming algorithm based on the satellite common-view system to improve the accuracy and frequency stability of rubidium clock.This thesis focuses on the taming technique of rubidium clock based on satellite common-view system.Compared with the time and frequency traceability system which traces the rubidium clock to GNSS,the common view method can not only eliminate the common ionospheric and tropospheric delay errors in the transmission path to a certain extent,but also offset the influence of satellite clock errors.The main research contents of this thesis are as follows:1.The basic principle of remote time and frequency traceability is studied,and the navigation satellite is used to transmit time and frequency over a long distance by means of common-view comparison to measure the bias of local rubidium clock.Then the rubidium clock bias is modeled and predicted to make up for the deficiency of the traditional rubidium clock taming algorithm that the differential unit has poor anti-noise interference ability and can not accurately predict the change of clock bias.By analyzing and comparing the modeling principles and prediction effects of several common clock bias prediction models.The clock bias prediction method combining grey model and quadratic polynomial model is selected according to the requirements for prediction error,and the influence of modeling sequence length on prediction accuracy is explored at the same time.2.A rubidium clock taming algorithm based on modified grey model and proportion integral differential(PID)control is proposed by combining the clock bias prediction model with PID control algorithm.By establishing modified grey model based on the original rubidium clock bias,the interference of satellite measurement noise and atomic clock noise can be suppressed,and the accuracy of clock bias prediction can be improved,thus improving the stability of PID taming results.By optimizing the PID control coefficient and shortening the PID control interval,the rubidium clock signal can closely track the change of the reference signal,so as to effectively improve the rubidium clock taming accuracy.The research results show that the rubidium clock taming algorithm proposed in this thesis can not only improve the rubidium clock taming accuracy to less than 0.2ns,but also improve the frequency stability of rubidium clock by four orders of magnitude,which provides the possibility to improve the performance of remote time and frequency traceability system.