In this activity we propose the development of new data compression algorithms and their software implementations for radiofrequency data (RF) and multi-band images. These will be built upon existing algorithms of the FAPEC data compression software, which has proven to be an efficient, reliable and versatile solution for cubesats. Actually, FAPEC, in its image compression configuration, is already being used onboard OPS-SAT, and one of its tailored options is being used in a commercial satellite constellation to compress RF data from their payloads. Starting from this existing framework reduces the risk of this activity, but our developments will be designed and implemented allowing for a separate execution from FAPEC. We will identify the limitations of these options, investigate existing approaches to overcome them, and if necessary, design brand new algorithms to provide the best possible results, at least for the OPS-SAT case. Remarkably, we intend to significantly improve the ratios achieved on RF data (in I/Q format), eventually focusing on just a small set of modulations. Smart lossy compression will be sought, aiming at adaptive losses depending on the noise level, as well as a fixed-rate lossy option. Regarding image compression, it will be thoroughly revised to also improve ratios, including better spatial-spectral decorrelators as well as inter-frame decorrelators for video. Fixed-rate lossy compression will be redesigned to allow for better quality levels at both small and large scale in the images. Implementations will be optimized aiming at high-throughput compression. The final goal of this activity is to allow for real-time in-orbit compression of data from demanding applications, such as long RF acquisitions or video. It should open new possibilities for cubesats, like large-scale spectrum utilisation analysis, studies of the upper ionosphere, entertainment and outreach, study of fast weather transients, or space hazards analysis.
