Title :
Data compression and VLSI implementation neuroprocessor
Author_Institution :
Jet Propulsion Lab., California Inst. of Technol., Pasadena, CA, USA
Abstract :
VLSI data compression research has been motivated by the needs of high-speed high-performance data compression and inspired by the VLSI technologies. An integrated data compression system has been proposed to provide adaptive multi-mode data compression for an advanced multi-instrument spacecraft payload system that has various source data. It combines a high-ratio lossy data compressor with a lossless data compressor to provide various compression schemes. The embedded lossy compressor is a high-speed neuroprocessor for adaptive vector quantization based upon a frequency-sensitive self-organization neural algorithm. The embedded lossless compressor is a high-speed pipelined processor design based on the Rice algorithm
Keywords :
application specific integrated circuits; data compression; image coding; mixed analogue-digital integrated circuits; neural chips; neural net architecture; parallel architectures; pipeline processing; self-organising feature maps; vector quantisation; very high speed integrated circuits; Rice algorithm; VLSI implementation; adaptive multi-mode data compression; adaptive vector quantization; data compression; frequency-sensitive; high-performance; high-ratio lossy data compressor; high-speed; high-speed neuroprocessor; image compression; integrated data compression system; lossless data compressor; mixed signal VLSI; multi-instrument spacecraft payload system; neuroprocessor; pipelined processor design; prototype neural processor chip; self-organization neural algorithm; tutorial; Algorithm design and analysis; Computer architecture; Data compression; Data engineering; Image coding; Image storage; Instruments; Space vehicles; Very large scale integration; Video compression;
Conference_Titel :
ASIC Conference and Exhibit, 1993. Proceedings., Sixth Annual IEEE International
Conference_Location :
Rochester, NY
Print_ISBN :
0-7803-1375-5
DOI :
10.1109/ASIC.1993.410846
