Numerical Optimization for the Simulation of Radio Occultation Events

The utilization of Global Navigation Satellite System (GNSS) to atmospheric radio occultation (RO) has been an active remote sensing method for exploring the Earth's atmosphere. The RO products, especially the atmospheric bending angles, have been widely used for climate study. To obtain these parameters, we need a signal model, and given the extensive distances traveled by RO signals, we need to establish a numerical method aimed at resolving the signal's path using ray tracing for real-time applications while optimizing computational cost. This paper focuses on computational optimization in the simulation of RO signals, primarily for events derived from GPS signals but applicable to any other GNSS system. We apply the Runge-Kutta-Fehlberg method for the signal path integration, with several spatial constraints, referred to as the planar ray parking (PRP) technique. Additionally, due to the nature of the problem, a shooting method is proposed. Moreover, we will propose a general RO simulation scheme that identifies blocks with potential for improvements in future work. Subsequently, we compare the performance of this new technique with a fourth-order Runge-Kutta (RK4) integrator, a commonly used algorithm for resolving trajectory of RO signals. The results demonstrate improved performance with negligible errors.