Data compression and SpaceWire

Abstract

Data compression has become an important part of a remote sensing instrument, with emphasis being placed on the total information content of data from an instrument rather than simple data quality. Future instruments with higher resolution and more channels (spectral bands for optical instruments and multi-polarisation, multi-frequency operation for synthetic aperture radar) will result in correspondingly higher data rates on-board the satellite. Data compression techniques can reduce the data rates while maintaining good data quality, however, the data compression hardware must be able to handle the high data rates from the sensor. This hardware must also be connected into the on-board data handling system without undue difficulty. This paper examines techniques for increasing the amount of data (bandwidth) that can be handled by data compression systems for SAR. Techniques for the compression of raw SAR data are introduced and the Block Adaptive Quantisation (BAQ) algorithm is described. Methods for improving the performance (sample rate) of BAQ are then considered. The need to balance data rates across the data handling chain and the effect that this has on the choice of data compression technique is discussed.