Evaluation of DME network capability using combination of rule-based model and gradient boosting regression

Abstract

The future development of aeronautical navigation foresees an infrastructure rationalization of radionavigation aids with the aim of maintaining only the Minimum Operational Network, which brings benefits in terms of operational cost savings, promotes sustainability and optimal use of the radio spectrum. To ensure that the necessary navigation performance is preserved, the Distance Measuring Equipment (DME) plays a significant role, as DME/DME navigation is a short-term contingency solution when Global Navigation Satellite System is unavailable. Therefore, new DME ground stations are put into operation even though other navigation aids are being decommissioned at the same time. This paper addresses a question of possible DME network rationalization by developing a software model using a combination of a rule-based model, approximating of airborne DME interrogators interacting with DME ground transponders, with the implementation of the Gradient Boosting Regression to predict load of DME ground stations. The model is validated by comparing the results with the real load data obtained from an Air Navigation Service Provider. Several test cases are performed to evaluate the capability of the European DME network, simulating a reduction in the number of en-route DME stations and increases in air traffic using clustering methods. The results show that the ground station load limit was rarely reached, demonstrating the robustness and the potential for rationalization of the DME infrastructure.