Algorithm for Computation of Troposphere Delay for Navigation and RADAR Stations

Abstract

Troposphere delay is one of the crucial factors of the atmospheric delays of the signal. The troposphere delay directly affects the line-of-sight error of the pseudo range measurements. A generalized model has been used to compute troposphere delay at a specific epoch. This model contains hydrostatic and non-hydrostatic components along with its gradient. The standard Saastamoinen model of troposphere delay can be derived from this generalized model. An algorithm is developed to compute the troposphere delay along with met sensor data using the generalized model. The met sensor data are obtained using three approaches. The first approach is the empirical model, the second approach is grid-based and the last one is real data. Results were obtained with a generalized troposphere model using all three approaches of met sensor data for the IGS (International GNSS Service) ground station. Results compared with the precise troposphere delay and found to be in good agreement. The accuracy is found to be within 28 cm. It can be concluded that the generalized troposphere is reliable and can be used for the correction in the line-of-sight error of the pseudo range measurement. A sensitivity analysis has been performed for troposphere delay to find the sensitive parameters. It has been found that the water vapor decrease factor is the most sensitive parameter among met sensor parameters (i.e., pressure, temperature, water vapor pressure), mean temperature and water vapor decrease factor. The water vapor pressure is the most sensitive among the met sensor parameters. The temperature is the least sensitive among all considered parameters.  Also, the present algorithm is used to compute the troposphere delay for NavIC (Navigation with Indian Constellation) system and RADAR (Radio Detection And Ranging) angle measurements. Finally, Troposphere delay is computed for real met sensor data (i.e., Approach 3) and compared with the results of the other two approaches. It has been found that due to uncertainty in the met sensor data, differences vary up to meter level.