Data compression/decompression method and apparatus

• US 5,140,321 A
• Filed: 09/04/1991
• Issued: 08/18/1992
• Est. Priority Date: 09/04/1991
• Status: Expired due to Term

• Current Assignee: PTC Inc.
• Original Assignee: Prime Computer, Inc. (DR Holdings, Inc. of Delaware)
• Inventors: Jung, Robert K.
• Primary Examiner(s): Logan, Sharon D.
• Application Number:
  US07/755,027
• Time in Patent Office: 349 Days
• Field of Search: 341/55, 341/50, 341/106, 341/95
• US Class Current: 341/55
• CPC Class Codes: H03M 7/3086 employing a sliding window, e.g. LZ77

Abstract

A method and apparatus for compressing digital data uses data which has been previously compressed as a dictionary of substrings which may be replaced in an input data stream. The method and apparatus uses a hash table to take advantage of principles of locality and probability to solve the maximal matching substring problem inherent in this type of compressing apparatus, most of the time. The hash table consists of first-in, first-out (FIFO) collision chains of fixed, uniform numbers of pointers to substrings of data already compressed which potentially match an input substring. A companion decompressing method and apparatus receives compressed data from the compressing apparatus and expand that data back to its original form.

Claims

1. A system for compressing digital input data, said input data being divisible into symbols, comprising:

a memory for holding at least first and second sequences of symbols from said input data, said first sequence having a prefix substring including fewer symbols than said first sequence;

means for receiving said first and second sequences of symbols from said memory and comparing said first and second sequences;

means for receiving said prefix substring from said memory and computing a hash value therefrom;

a hash table for receiving said hash value and providing a pointer to said second sequence of symbols, said hash table including a plurality of first in, first-out (FIFO) collision chains, having identical numbers of pointers to locations in said memory; and

means for generating compressed output data responsive to said means for receiving said first and second sequences of symbols from said memory.

2. A system as claimed in claim 1, wherein said memory further comprises:

a first buffer for holding said first sequence; and

a second buffer for holding at least one said second sequence.

3. A system as claimed in claim 1, wherein said compressed output data includes an information portion selected from a group consisting of a pointer to one said second sequence, a partial pointer to one said second sequence, and a first symbol of said first sequence;

and a flag portion for identifying which member of said group is selected; and

said means for generating compressed output data further comprises;

means for generating said information portion; and

means for generating said flag portion.

4. A system as claimed in claim 3, wherein said pointer comprises:

an offset for identifying a first symbol of said second sequence; and

a length over which said first sequence and said second sequence match.

5. A system as claimed in claim 3, wherein said partial pointer comprises:

a length over which said first sequence and said second sequence match; and

without including an offset for identifying a first symbol of said second sequence.

6. A system for processing digital input data, said input data being divisible into sequences of bits representing symbols, comprising:

a memory for holding at least first and second sequences of symbols from said input data, said first sequence having a prefix substring of symbols of a fixed length;

means for receiving said first and second sequences of symbols from said memory and determining a length over which said first and second sequences match;

means for receiving said prefix substring and computing a hash value therefrom;

a hash table for receiving said hash value and providing a pointer to said second sequence of symbol signals, said hash table including a plurality of first in, first-out (FIFO) collision chains, having identical numbers of pointers to locations in said memory;

means for generating compressed data responsive to said means for receiving said first and second sequences of symbols from said memory;

means for receiving said compressed data;

means for decompressing said compressed data received.

7. A system as claimed in claim 6, wherein said compressed data includes an information portion selected from a group consisting of a pointer to one said second sequence, a partial pointer to one said second sequence, and a first symbol of said first sequence;

and a flag portion for identifying which member of said group is selected; and

said means for generating compressed data further comprises;

means for generating said information portion; and

means for generating said flag portion.

8. A system as claimed in claim 7, wherein said means for decompressing said compressed data further comprises:

means for decoding said flag portion; and

means responsive to said means for decoding said flag portion, for expanding said information portion into a decompressed data sequence identical to said input data.

9. A system as claimed in claim 8, wherein said means for expanding said information portion further comprises:

a memory for receiving sequences of decompressed data; and

means for identifying in said memory for receiving decompressed data sequences pointed to by said pointers and said partial pointers; and

means for copying said sequences identified by said means for identifying to new locations in said memory after a most recently decompressed symbol.

10. A system as claimed in claim 6, wherein said means for generating compressed data further comprises:

means for transmitting said compressed data to a remote location.

11. A system as claimed in claim 10, wherein said means for receiving said compressed data is disposed at said remote location.

12. A method for processing digital input data, said input data being divisible into symbols, comprising the steps of:

receiving input data, including a first sequence of symbols having a prefix substring including fewer symbols than said first sequence;

computing a hash value of said prefix substring;

obtaining a pointer to a second sequence of symbols in said input data, from a hash table including a plurality of first in, first-out collision chains, having identical numbers of pointers to sequences of symbols;

comparing said first and second sequences; and

generating compressed data output based on said comparison of said first and second sequences.

13. A method as claimed in claim 12, wherein the step of generating further comprises the steps of:

outputting information data selected from a group consisting of a pointer to one said second sequence, a partial pointer to one said second sequence, and a first symbol of said first sequence; and

outputting flag data for identifying which member of said group is selected.

14. A method as claimed in claim 12, further comprising the step of:

decompressing said compressed data.