HR: 0800h
AN: SM21A-0349    [Abstracts]
TI: Adaptive Data Compression Module for Auroral Images
AU: * Kim, H
EM: hyojin.kim@unh.edu
AF: University of New Hampshire Space Science Center, Morse Hall, Durham, NH 03824 United States
AU: Lessard, M R
EM: marc.lessard@unh.edu
AF: University of New Hampshire Space Science Center, Morse Hall, Durham, NH 03824 United States
AU: Riley, P
EM: paul.riley@unh.edu
AF: University of New Hampshire Space Science Center, Morse Hall, Durham, NH 03824 United States
AB: Due to their inherent finite bandwidth, satellites and sounding rockets can only provide data at a limited rate. The same is true for ground-based observatories that transmit their data via satellite. The situation can be improved quite dramatically using appropriate data compression schemes. For the particular case of the upcoming Rocket Observations of Pulsating Aurora (ROPA) mission, which incorporates two auroral imagers, development of an image compression module on board reduces the needed resources, both in terms of mass on the payload as well as cost. For this reason, image compression software is being developed for ROPA, although the technique will clearly be useful for other applications (i.e., satellite and ground-based observatories). The main compression algorithm, similar to the lossless JPEG-LS algorithm, is based on the idea that the auroral images tend to display continuous-tones, where the neighboring pixels are approximately predictable. We concentrate on simply modifying the algorithm and optimizing the implementation to be fairly suitable for hardware characteristics of the FPGA (Field Programmable Gate Array) in the imagers. A pre-compression process to split an entire image into low-entropy small-pixel blocks enables the algorithm to be adapted to varying entropy. Initial tests indicate compression ratios of approximately 50% for 8-bit ground-based auroral allsky images. The expectation is that for the 10-bit image files on ROPA, compression ratios of 65% may be achievable. The technique appears promising and may also be useful for time series data.
DE: 2400 IONOSPHERE (6929)
DE: 2407 Auroral ionosphere (2704)
DE: 2494 Instruments and techniques
DE: 2704 Auroral phenomena (2407)
DE: 9820 Techniques applicable in three or more fields
SC: SPA-Magnetospheric Physics [SM]
MN: Fall Meeting 2005