Study of ARFTIS reconstruction model on GPU

There are huge calculations on reconstruct processing of all-reflective Fourier transform imaging spectrometer (ARFTIS) data. Common algorithms often take long processing time. The common unified device architecture (CUDA) is a fundamentally new programming approach making use of the unified shader design of the most current graphics processing units (GPUs) from NVIDIA. We research GPU develop platform and compare with CPU algorithmic structure. The results show that the CUDA may accelerate multiplications and additions computing significantly, that are the great mass of ARFTIS reconstruction. The CUDA programming interface allows implementing an algorithm using standard C language. Farther, we give some suggestions to optimize reconstruction algorithms on GPU.