Research On The Modeling Techniques Of GPS Receiver Clock Model

At present, GPS has the most mature application in the four satellite navigationsystems. Of all the current means of time services, satellite time service is the mostwidely used timing technique. When using GPS to provide time service, time biasdata of the receiver needs to be calculated and modeled to forecast the clock bias ofthe future time. Therefore, it is of great importance to study the modeling technique ofGPS receiver clock model.In order to make a more scientific and accurate forecast of the future status ofreceiver clock bias, the analysis and forecast method of receiver clock bias presentedin this paper adopted not only the widely used quadratic polynomial model, but alsokalman model and grey theory model, each of which got a detailed theoreticaldescription and application prospect discussion. Besides, the solutions of the relevantparameters were discussed as well.Meanwhile, in order to prevent the bad influence that the old receiver clock biasdata brings to the clock bias forecast, this paper adopts sliding mode to continuouslyimport new data and eliminate the old data to ensure the parameter precision of themodel, which allows the clock bias forecast to make a better corresponding changeaccording to current tendency. The accuracy of clock bias forecast is mainly affectedby two aspects, which are point number of the clock bias modeling and time intervalof the forecast model. Therefore, this paper establishes these three models by usingsimulation data of different window length, which are1minute,3minutes and5minutes, to forecast the receiver clock bias of the subsequent1second,30secondsand60seconds via sliding mode.Using the satellite data obtained by receiver to simulate these three clock biasforecast models, the results show that when using a small amount data to establishforecast model, quadratic polynomial model and kalman model are only suitable forshort-term clock bias forecast, because they have poor long-term forecast accuracy.While when it comes to gray theory model, both the short-term and long-term forecastprecision are better that the other two models. When using large amount of data toestablish forecast models, there is no distinct difference among the forecast precisionof these three models. By taking the complexity and accuracy of these three models,and the calculation capacity of the receiver into consideration, this paper concluded that the most suitable forecast model for practical application is grey theory model.On the basis of the above research conclusions, this paper further analyzes andsimulates how to use grey theory forecast model to assist the receiver to achievethree-dimensional positioning under the condition that only3satellite signal can bereceived. The simulation results show that this method can assist the receiver toachieve three-dimensional positioning within a short time, which greatly improves thecontinuity of positioning in harsh conditions.