Filtering Enhanced Traffic Management System (ETMS) Altitude Data
Name
Palacios-2013-Filtering Enhanced T.pdf
Size
598.65 KB
Format
Adobe PDF
Checksum (MD5)
8e12eed8076de9e3d86cd45ec3417b97
Author(s)
Palacios, Rafael
Hansman, Robert J., Jr.
Hansman, Robert J., Jr.
Date Issued
September 2013
Journal
Metrology and Measurement Systems
Publisher
Versita (Central European Science Journals)
Citation
Palacios, Rafael, and R. John Hansman. “Filtering Enhanced Traffic Management System (ETMS) Altitude Data.” Metrology and Measurement Systems 20, no. 3 (n.d.). © 2013 Polish Academy of Sciences
Version
Final published version
Abstract
Enhanced Traffic Management System (ETMS) stores all the information gathered by the Federal Aviation Administration (FAA) from aircraft flying in the US airspace. The data stored from each flight includes the 4D trajectory (latitude, longitude, altitude and timestamp), radar data and flight plan information. Unfortunately, there is a data quality problem in the vertical channel and the altitude component of the trajectories contains some isolated samples in which a wrong value was stored. Overall, the data is generally accurate and it was found that only 0.3% of the altitude values were incorrect, however the impact of these erroneous data in some analyses could be important, motivating the development of a filtering procedure. The approach developed for filtering ETMS altitude data includes some specific algorithms for problems found in this particular dataset, and a novel filter to correct isolated bad samples (named Despeckle filter). As a result, all altitude errors were eliminated in 99.7% of the flights affected by noise, while preserving the original values of the samples without bad data. The algorithm presented in this paper attains better results than standard filters such as the median filter, and it could be applied to any signal affected by noise in the form of spikes.
MIT Department
Massachusetts Institute of Technology. Department of Aeronautics and Astronautics
Terms of Use
Creative Commons Attribution-Non-Commercial-NoDerivs license
Persistent DSpace Link
DOI of Published Version
https://doi.org/10.2478/mms-2013-0039
