Set Theoretic Processor Pattern Recognition Network

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current pattern recognition technologies face challenges in achieving simultaneous high acuity, throughput, scalability, and cost-effectiveness in recognizing specified patterns in data streams, particularly due to the combinatorial explosion problem and limitations in handling complex patterns and equivalence classes.

Innovation Solution

A general purpose set theoretic processor with a recognition network, aggregation network, and local clock and sequencer, utilizing a state-change routing matrix, detection cells, pattern memory, and threshold logic to process complex patterns and equivalence classes in a single semiconductor device, enabling efficient recognition of relevant data entities.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional pattern recognition technologies are used to recognize complex patterns, then pattern recognition capability is improved, but device complexity increases due to combinatorial explosion

Engineering Contradiction:
Improvepattern recognition capabilityVSAvoiddevice complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent divides the pattern recognition task into multiple independent detection cells, each responsible for detecting specific features or substrings. This segmentation allows the system to handle complex patterns by breaking them down into manageable units, reducing the combinatorial explosion that would occur if a single monolithic approach were used.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Multiple detection cells are merged into a parallel array structure that operates simultaneously. The detection cells are connected through a routing matrix that combines their outputs to recognize complete patterns. This merging approach achieves high pattern recognition capability while maintaining linear scalability rather than exponential complexity growth.

Inventive Principle:
Principle #5Merging (Combining)

2Measurement precision

If pattern recognition systems process complex patterns with many features, then measurement precision is improved, but throughput decreases

Engineering Contradiction:
Improvepattern recognition accuracyVSAvoidthroughput
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The pattern recognition system is segmented into multiple detection cells that independently process different features or substrings in parallel. Each cell operates on the same input data simultaneously, allowing the system to maintain high throughput while achieving high measurement precision through the combined results of all cells.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces a temporal dimension by processing patterns in sequential passes or cycles. Multiple patterns or features are recognized in parallel during each pass, and the results are aggregated over time. This dimensional approach allows the system to maintain constant throughput while improving precision by examining more features over multiple cycles.

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

3Adaptability or versatility

If pattern recognition systems are designed to handle large numbers of features and terms, then adaptability is improved, but manufacturing cost increases

Engineering Contradiction:
Improvepattern handling capabilityVSAvoidmanufacturing cost
Core Design Contradiction:
Adaptability or versatilityVSEase of manufacture

Solution Approach 1:

The detection cells are designed with universal functionality to detect various types of patterns and features using the same basic hardware structure. The routing matrix and detection cells can be configured to recognize different patterns by loading different reference patterns into memory, eliminating the need for custom hardware for each pattern type and reducing manufacturing costs.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

Instead of manufacturing custom hardware for each pattern recognition task, the system uses multiple copies of the same detection cell architecture. These identical cells are instantiated in parallel and configured through software or memory loading to handle different patterns. This copying approach maintains high adaptability while significantly reducing manufacturing complexity and cost.

Inventive Principle:
Principle #26Copying

4Adaptability or versatility

If pattern recognition systems are scaled up to handle larger datasets and more complex queries, then adaptability is improved, but device complexity increases

Engineering Contradiction:
Improvequery handling capabilityVSAvoidsystem complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system is divided into multiple detection cells that can be independently configured and operated. This segmentation allows the system to scale by adding more cells rather than increasing the complexity of individual cells. Each cell maintains a simple structure while the collective system handles increasingly complex queries through parallel processing.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent implements detection cells that may perform more operations than strictly necessary for a single simple query. By designing cells with excess capability and using them in parallel, the system can handle complex queries efficiently without requiring each cell to be overly complex. The partial results from multiple cells are combined to achieve the final query result.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentUS7487131B2General Purpose set theoretic processor
Publication Date: 2009.02.03 HARRIS CURTIS L
  • US7487131B2 patent drawing
  • US7487131B2 patent drawing
  • US7487131B2 patent drawing

AI summary

A general purpose processor architecture (methods and apparatuses) that can discern all subsets of a serial data stream which fulfill an arbitrarily complex reference pattern. The invention comprises an ordered set of Detection Cells conditionally interconnected according to the reference pattern and operationally controlling one another's states through the network. The invention preferably includes a Host Interface to enable reporting of Results from a search session as well as the input and control of reference patterns and source data.