Dismiss
The InnovationQ application will be updated on Sunday, May 31st from 10am-noon ET. You may experience brief service interruptions during that time.
Browse Prior Art Database

Adaptive Context Generation For Data Compression

IP.com Disclosure Number: IPCOM000050372D
Original Publication Date: 1982-Oct-01
Included in the Prior Art Database: 2005-Feb-10

Publishing Venue

IBM

Related People

Authors:
Rissanen, J [+details]

Abstract

This invention relates to a method for adaptive context generation for data compression and the like, comprising the steps of (1) suitably ordering past symbol strings to form a selective structure and (2) growing contexts within the selected structure in a nested manner responsive to successive symbols, thereby minimizing memory to actually occurring contexts. The main problem in designing a data compression system is to select for each symbol in the string to be compressed its context, which means a certain collection of its adjacent past symbols. This process together with gathering the occurrence counts of the symbol at its context is known as modeling an information source for the string.