Study And Modeling On Temporal Variation Property Of Low-frequency Ground Wave Propagation Delay In Land-based Long-wave Navigation Timing System

The land-based long-wave navigation timing system is an indispensable part of complete positioning,navigation and timing(PNT)system.However,in order for the system to achieve high-quality PNT service,it is crucial to obtain accurate and reliable low-frequency ground wave propagation delay.Due to complex changes in the electrical characteristics of the propagation path,low-frequency ground wave propagation delay exhibits obvious temporal variation property.Therefore,it is of great significance to study the temporal variation property of low-frequency ground wave propagation delay and its prediction methods.Based on the realistic demand for high-precision prediction of low-frequency ground wave propagation delay,this article focuses on the temporal variation property of low-frequency ground wave propagation delay.Aiming at the problem of propagation delay prediction under complex and long distance,a modeling method that considers multi-point and multi-weather on the propagation path and uses generalized regression(GR)neural network is proposed,which effectively improves the prediction accuracy of propagation delay.The specific content is as follows:(1)Based on the theoretical prediction method,the mechanism of various time-varying factors in the stratum and atmosphere on the temporal variation property of propagation delay,as well as the law and degree of influence on the propagation delay are analyzed.The results show that there is an inevitable connection between meteorological factors and propagation delay,and for a propagation distance of 1000km,the variation in propagation delay caused by meteorological changes can reach the order of hundreds of nanoseconds or even the order of microseconds.(2)Based on long-term monitoring data,the time-varying law of low-frequency ground wave propagation delay and its correlation with meteorological factors are analyzed.The following conclusions are obtained:First of all,the propagation delay varies with seasons,day and night,and its fluctuation amplitude increases with the increase of propagation distance;In addition,compared with the simple short path,for complex long path,the time-varying degree of propagation delay is the result of the combined effect of meteorological factors at various locations on the entire path;Finally,the meteorology of a certain location on the complex long propagation path will not be sufficient to reflect the meteorology on the whole propagation path,especially when the meteorological differences along the propagation path are huge.Therefore,the weather at more points along the propagation path needs to be considered.(3)The propagation delay prediction model suitable for complex long path is established.Aiming at the complex long-distance situation,taking the all-land path and the sea-land path as examples,propagation delay prediction models that consider single-point or multi-point weather are established by using backward propagation(BP)neural network and GR neural network.The results show that the modeling method that considers the multi-point and multi-weather on the propagation path and uses the GR neural network is more effective.In addition,this modeling method is also universally applicable.The research results of this paper can provide certain reference value for the research of high-performance land-based long-wave navigation timing system.