A Comparison of LAAS Error Bounding Concepts

Curtis A. Shively

Abstract:	The FAA™s Local Area Augmentation System (LAAS)broadcasts a parameter ó pr_gnd to describe the errors in the differential corrections due to the ground facility. An aircraft using LAAS computes an upper bound with high probability on the resulting error in the position domain based on ó pr_gnd assuming that the errors are Gaussian distributed. This paper compares five bounding methods in terms of assumptions and resulting performance. Assumptions regarding the tails of the error distribution range from a Gaussian model to an exponential model. Performance is compared in terms of the factor by which the estimated value of ó pr_gnd must be inflated before broadcast to ensure the bound is provided with known confidence. For fixed, desired confidence the inflation factor varies with the number of independent samples used to estimate ó pr_gnd. Results show that using the same number of samples, methods assuming Gaussian tails give significantly smaller inflation factors than does the exponential tail method. Between these two extremes, is a more recently conceived method that assumes the error distribution is a mixture of Gaussian distributions with different standard deviations. The Gaussian mixture method gives inflation factors that are smaller than for the exponential tail method and may most closely correspond to the manner in which errors are present in real data.
Published in:	Proceedings of the 2001 National Technical Meeting of The Institute of Navigation January 22 - 24, 2001 Westin Long Beach Hotel Long Beach, CA
Pages:	501 - 511
Cite this article:	Shively, Curtis A., "A Comparison of LAAS Error Bounding Concepts," Proceedings of the 2001 National Technical Meeting of The Institute of Navigation, Long Beach, CA, January 2001, pp. 501-511.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In