Tomasi Maurizio

Abstract Reference: 30748
Identifier: P6.25
Presentation: Poster presentation
Key Theme: 6 Python in Astronomy

Compression of smooth one-dimensional data series using "polycomp"

Authors:
Tomasi Maurizio

Data compression is increasingly important in astrophysics, as the amount of data acquired by modern experiments often needs hundreds of terabytes for the storage of raw data. In this talk I will present a few usage cases of the C/Python library "polycomp", a library to compress smooth one-dimensional data whose error is either zero or negligible. One of the algorithms implemented by "polycomp" combines the advantages of polynomial least-squares fitting and the properties of the discrete Chebyshev transform. This algorithm can lead to compression ratios larger than 10 in a number of realistic cases. I will show a few examples of datasets that can be easily  compressed using this approach, namely (1) spacecraft attitude information, and (2) timelines of pointing information for a realistic all-sky survey experiment.