A neural network based algorithm for precise transformation between GPS height and pressure altitude

This paper proposes a neural network based algorithm for precise transformation between GPS height and pressure altitude. In order to accomplish the transformation, the algorithm adopts geometric height referenced to MSL as a mid-transfer parameter by which the entire process is divided into two components: the transformation between two reference surfaces and the transformation between two height measurements. The algorithm can be used in applications that require high accuracy position information, such as the monitoring of aircraft height keeping performance in reduced vertical separation minimum (RVSM) airspace and precise positioning for aerostat.