Content-Addressable Memory Parallel Data Address Comparison

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current information processing systems, particularly those based on von Neumann architecture, face inefficiencies in searching and detecting information due to sequential processing, leading to high search times and reduced detection accuracy, especially when dealing with vast amounts of data.

Innovation Solution

A computer-readable memory that enables parallel comparison of data and address relationships, allowing for fast and accurate information detection without sequential address processing, utilizing a content-addressable memory with dual parallel logic operations to reduce the number of comparison steps and enhance information refinement.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If sequential address processing is used in von Neumann architecture, then device complexity is reduced, but information detection speed and accuracy deteriorate

Engineering Contradiction:
Improveinformation detection speedVSAvoidmemory processing complexity
Core Design Contradiction:
SpeedVSDevice complexity

Solution Approach 1:

The patent segments the memory system into multiple independent memory units, each capable of autonomous parallel comparison operations. This segmentation enables simultaneous processing of multiple data items without sequential address processing, thereby improving detection speed while distributing system complexity across modular units.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent transitions from one-dimensional sequential address processing to multi-dimensional parallel comparison by enabling simultaneous comparison across multiple memory units using different comparison keys. This dimensional expansion allows information detection to occur in parallel across multiple dimensions, dramatically increasing speed without proportionally increasing complexity.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Speed

If the number of comparison combinations is reduced to speed up information detection, then information detection speed improves, but detection accuracy deteriorates

Engineering Contradiction:
Improveinformation detection speedVSAvoidinformation detection accuracy
Core Design Contradiction:
SpeedVSMeasurement precision

Solution Approach 1:

The patent implements a feedback mechanism where comparison results from multiple parallel operations are aggregated and refined through iterative processes. The system uses feedback loops to verify and refine detection results, ensuring high accuracy even when the number of comparison combinations is reduced, thereby resolving the trade-off between speed and accuracy.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The patent merges multiple comparison results from parallel memory units through logical operations to produce a unified detection outcome. By combining results from multiple independent comparisons, the system achieves both speed (through parallel processing) and accuracy (through result aggregation and verification).

Inventive Principle:
Principle #5Merging (Combining)

Data Source

PatentEP2538348B1Memory having information refinement detection function, information detection method using memory, device including memory, information detection method, method for using memory, and memory address comparison circuit
Publication Date: 2022.06.15 INOUE KATSUMI
  • EP2538348B1 patent drawingFigure 1
  • EP2538348B1 patent drawingFigure 2
  • EP2538348B1 patent drawingFigure 3

AI summary

There is provided an externally readable memory for storing information in each memory address, and this memory is provided with an information refinement detection function; this memory comprises: an input means for entering first input data for comparing data items stored in the memory and second input data for comparing addresses in the memory, wherein the first and second comparison data are externally; means for determining matches/mismatches of both data items stored in the memory and addresses of the memory according to both of the input data provided by the input means, and further performing logic operations on both of the match/mismatch determination results; and means for outputting addresses with positive results of the logic operations.This memory may be applicable in a broad range of fields including intelligent information search as well as artificial intelligence.