Method and apparatus with data compression and/or decompression
AssigneeSamsung Electronics Co., Ltd.
InventorsHyeongseok Yu, Deok Jae Oh, Seong Wook Park
Life of this patent
Estimating the term…
Abstract
A processor-implemented method including generating k sub-compressed data streams based on a compressed data stream for a plurality of symbols divided into a plurality of k blocks and count information for each of the plurality of k blocks, generating k sub-symbols by processing each of the k sub-compressed data streams using k decoding engines, metadata about the compressed data stream, and generating an output data stream corresponding to the plurality of symbols based on the k sub-symbols.
Claims20
A processor-implemented method, the method comprising:
- obtaining target data comprising a compressed data stream for a plurality of symbols divided into a plurality of k blocks, metadata about the compressed data stream, and count information for each of the plurality of k blocks;
- storing the compressed data stream in an input buffer;
- generating k sub-compressed data streams based on the count information and the compressed data stream stored in the input buffer;
- generating k sub-symbols by processing each of the k sub-compressed data streams using k decoding engines;
- and
- generating an output decoded data stream corresponding to the plurality of symbols based on the k sub-symbols.
An electronic apparatus, the apparatus comprising:
- a processor configured to execute instructions;
- and
- a memory storing the instructions, wherein execution of the instructions by the processor configures the processor to:
- obtain target data comprising a compressed data stream for a plurality of symbols divided into a plurality of k blocks, metadata about the compressed data stream, and count information for each of the plurality of k blocks;
- store the compressed data stream in an input buffer;
- generate k sub-compressed data streams based on the count information and the compressed data stream stored in the input buffer;
- generate k sub-symbols by processing each of the k sub-compressed data streams using k decoding engines;
- and
- generate a decoded output data stream corresponding to the plurality of symbols based on the k sub-symbols.
A processor-implemented method, the method comprising:
- dividing a plurality of symbols into a plurality of k blocks;
- generating first codewords by compressing symbols of a first block among the plurality of k blocks;
- generating first partial metadata for the compressing;
- generating first count information indicating a length of the first codewords;
- and
- generating compressed data representing the plurality of symbols based on the first codewords, the first partial metadata, and the first count information.
Dependencies are inferred from the claim text.