SAR Satellite Data Reduction using Dynamic 4 Path – Block Gain Tree Structured Vector Quantization
- Paper number
IAC-15,B1,4,5,x27450
- Author
Dr. Hyeon-Cheol Lee, Korea Aerospace Research Institute (KARI), Korea, Republic of
- Coauthor
Dr. Eun Su Kang, Korea Aerospace Research Institute (KARI), Korea, Republic of
- Coauthor
Dr. Sang Soon Yong, Korea Aerospace Research Institute (KARI), Korea, Republic of
- Year
2015
- Abstract
A SAR payload used in Satellite sends a ground station its SAR raw data by wireless datalink for post-processing, then the efforts of reducing its SAR raw data have been studied much. Vector Quantization among these efforts is used much with block adaptive quantization. Due to heavy data search load of Vector Quantization between codebook and image data, Tree Structured Vector Quantization(TSVQ) is introduced instead of full-search-codebook. In this paper, Dynamic 4 Path - Block Gain Tree Structured Vector Quantization(D4P-BGTSVQ) which has 4-path tree structures with block gain is introduced instead of conventional 2-path TSVQ selectable at each tree node, and we suggest the best solution by comparing SNR and calculation load.
- Abstract document
- Manuscript document
(absent)