Computation system and method using hamming distance

• US 5,216,750 A
• Filed: 02/28/1991
• Issued: 06/01/1993
• Est. Priority Date: 06/07/1990
• Status: Expired due to Term

• Current Assignee: Texas Instruments, Inc.
• Original Assignee: Texas Instruments, Inc.
• Inventors: Smith, Derek J.
• Primary Examiner(s): MacDonald, Allen R.
• Application Number:
  US07/662,041
• Time in Patent Office: 824 Days
• Field of Search: 364/513, 382/14, 382/15, 395/24, 395/27
• US Class Current: 706/18
• CPC Class Codes: G06F 18/2135 based on approximation criteria, e.g. principal component analysis
• CPC Class Codes: G06N 3/044 Recurrent networks, e.g. Hopfield networks
• CPC Class Codes: G06T 9/002 using neural networks
• CPC Class Codes: H04N 19/00 Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
• CPC Class Codes: H04N 19/60 using transform coding
• CPC Class Codes: H04N 19/94 Vector quantisation

Abstract

Preferred embodiments include systems with neural network processors (58) having input encoders (56) that encode integers as binary vectors so that close integers encode as close binary vectors by requiring adjacent integers have encoded binary vectors that differ in a fixed fraction of their bits.

Claims

1. A method of processing information comprising:

providing a set of information signals;

receiving one of said set of information signals at an input of an encoder;

generating a plurality of series of electrical signals in said encoder, each series associated with a corresponding one of said set of information signals, said generating step comprising the steps of;

setting a selected number of said electrical signals in said series associated with the first information signal to a selected state; and

for each of the remaining said series, setting said electrical signals to the states of the preceding series and then randomly changing a number of said electrical signals while maintaining the remaining electrical signals such that the Hamming distance between any two series is, within a selected probability, less than a selected value if the two corresponding information signals are within a given closeness;

providing said electrical signals associated with said one of said set of information signals to an input of a neural network; and

processing said electrical signals within said neural network.

2. The method of claim 1, wherein said generating step comprises generating a lookup table.

3. The method of claim 1, wherein said generating step comprises generating the contents of a memory.

4. The method of claim 1, wherein said step of providing a set of information signals comprises:

viewing a physical image and creating an electrical signal with characteristics corresponding to the physical image;

digitizing said electrical signal in an analog-to-digital converter; and

performing a Fast Fourier Transform on said digitized electrical signal.

5. The method of claim 4, wherein said step of viewing a physical image comprises viewing a physical image through a television camera.

6. The method of claim 1, wherein said step of providing a set of information signals comprises providing sensor information and said step of processing said electrical signals within said neural network comprises classifying said sensor information.

7. The method of claim 1, wherein said step of generating a plurality of electrical signals comprises the step of generating two state electrical signals.

8. The method of claim 1, wherein said generating step is performed once and said steps of receiving, providing, and processing are performed more than once.

9. The method of claim 1, wherein said step of changing a number of said electrical signals comprises the step of changing the same number of electrical signals for each of said series.

10. The method of claim 1, wherein said Hamming distance is between the number one and thirty-five percent of the number of electrical signals.

11. The method of claim 1, wherein said Hamming distance is distributed with an average in the range between five percent of the number of electrical signals and twenty percent of the number of electrical signals.