| The Geosynchronous Imaging Fourier Transform Spectrometer (GIFTS) is an imaging interferometer currently being designed as part of NASA's New Millennium Program Earth Observing 3 (NMP EO-3) mission and is scheduled to be launched in 2005. This thesis investigates two different broad classes of data compression algorithms for GIFTS data; first, computationally simple differential-pulse-code-modulation (DPCM)-based algorithms, and second, an algorithm which employs the state-of-the-art Set Partitioning in Hierarchical Trees (SPIHT) image compression algorithm. Bit-level simulations of the two algorithms are performed on simulated GIFTS data-cubes and compression errors are reported in the interferogram domain, spectral radiance domain, and temperature and water vapor retrieval domains. The question of optimal bit allocation is also addressed and a genetic algorithm optimization of bit allocation is presented. |
