Wavelet Compression Performance of MMS/FPI Plasma Count Data with Plasma Environment. Issue 1 (29th January 2019)

• Record Type: Journal Article
• Authors: Barrie, A. C.; Smith, D. L.; Elkington, S. R.; Sternovsky, Z.; da Silva, D.; Giles, B. L.; Schiff, C.
• Is Part Of: Earth and space science. Volume 6:Issue 1(2019)
• Page Start: 116
• Page End: 135
• Publication Date: 2019-01-29
• Subjects: compression -- wavelet -- MMS -- FPI -- plasma; Space sciences -- Periodicals; Geophysics -- Periodicals; 500.5
• DOI: 10.1029/2018EA000430
• Languages: English
• ISSNs: 2333-5084

Abstract

Abstract : The Fast Plasma Investigation (Pollock et al., 2016, https://doi.org/10.1007/s11214-016-0245-4 ; FPI) onboard the Magnetospheric Multiscale mission (Burch, Moore, et al., 2016, https://doi.org/10.1007/s11214-015-0164-9 ; MMS) uses a discrete wavelet transform and bit plane encoder (DWT/BPE; Winterrowd et al., 2010, https://doi.org/10.1109/AERO.2010.5446664 ) for data compression. This is the first plasma spectrometer suite to use this method of compression and thus serves as a benchmark for future plasma spectrometers. Flight data from MMS confirm viability of this compression algorithm in large areas of the magnetosphere. Because much of the FPI data were compressed losslessly on orbit, this flight data can be used as seed data to investigate performance of the DWT/BPE‐based compression at increased rates of compression. In this study, data from several representative regions of the magnetosphere have been compressed to increasingly small sizes and the resulting error was analyzed. Wavelet‐based compression is shown to be effective in all regions of the magnetosphere and solar wind for plasma count data, with performance varying with local environment. Specifically, plasma distributions that are characterized by low temperature and/or low density are compressed better leading to excellent performance in plasma regions such as solar wind and the magnetosheath. In general, ion data are compressed better than electron data, primarily due to a higher drift velocity of ions relative to their thermal speed and lower count rates. DWT/BPE compression can therefore be recommended for future instruments measuring count data in Earth's magnetosphere and in solar wind. Plain Language Summary: Space science missions are using increasingly advanced instrumentation and are taking larger and larger amounts of data. As such, better forms of data compression are required. This work explores using a wavelet‐based data compression scheme to compress plasma count data. The results indicate that wavelet compression can store plasma count data in a volume roughly 10 times smaller than techniques used for previous space missions with low error introduced. Key Points: Wavelet compression can be used to compress plasma count data with an error of the order of Poisson noise Wavelet compression is more efficient in lower density, lower temperature plasmas Electron data compression is less efficient than ion data, and the efficiency varies according to the plasma regime explored