Arithmetic Device XNOR Logic Magnetic Stacked Bodies

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current arithmetic devices with magnetic elements require complex configurations to perform XNOR operations, which can be simplified to achieve more efficient and cost-effective operations.

Innovation Solution

The arithmetic device includes a first and second element with conductive members and stacked bodies, utilizing magnetic layers and counter magnetic layers to perform XNOR operations based on electrical resistance and potential changes, allowing for a simpler configuration and operation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If magnetic elements are used to perform XNOR operations, then computational functionality is achieved, but device complexity increases

Engineering Contradiction:
ImproveXNOR operation capabilityVSAvoidconfiguration complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent combines multiple functional elements (first element with first conductive member and first stacked body, second element with second conductive member and second stacked body) into a unified arithmetic device structure. The magnetic layers and counter magnetic layers are integrated with conductive members to form compact elements that perform XNOR operations through their combined magnetic and electrical properties, reducing overall device complexity while maintaining functionality.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The arithmetic device uses magnetic elements that can represent both data storage and computational functions. The first and second stacked bodies with magnetic layers serve dual purposes: storing input data states and participating in the XNOR computation process itself, eliminating the need for separate storage and computation units.

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

2Reliability

If complex configurations are used to perform XNOR operations with magnetic elements, then operational reliability is improved, but power consumption increases

Engineering Contradiction:
Improveoperation reliabilityVSAvoidpower consumption
Core Design Contradiction:
ReliabilityVSUse of energy by moving object

Solution Approach 1:

The magnetic elements in the patent utilize their inherent magnetic properties and resistance changes to automatically indicate computation results. The device leverages the natural physical behavior of magnetic layers and conductive members under applied voltages, eliminating the need for additional active control circuits or power-intensive readout mechanisms that would increase power consumption.

Inventive Principle:
Principle #25Self-service

3Device complexity

If simpler configurations are used for arithmetic operations, then device complexity is reduced, but measurement precision of electrical resistance and potential changes deteriorates

Engineering Contradiction:
Improveconfiguration simplicityVSAvoidelectrical resistance measurement precision
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The patent employs localized magnetic layers and conductive members with specific structural characteristics optimized for their measurement function. Each element's physical dimensions, material composition, and geometric arrangement are tailored to enhance the sensitivity and precision of electrical resistance and potential change measurements while keeping the overall device configuration simple.

Inventive Principle:
Principle #3Local quality

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

This configuration enables efficient XNOR operations with reduced complexity and power consumption, suitable for applications in edge computing and AI, particularly in Binary Neural Networks (BNNs).

Implementation Method 1

The first stacked body includes a first magnetic layer, and a first counter magnetic layer provided between the third portion and the first magnetic layer. The second stacked body includes a second magnetic layer, and a second counter magnetic layer provided between the sixth portion and the second magnetic layer.

Methodology Applied
Scientific EffectMagnetoresistance: Magnetoresistance

Data Source

PatentUS20220291898A1Arithmetic device
Publication Date: 2022.09.15 SP AITH LTD
  • US20220291898A1 patent drawing
  • US20220291898A1 patent drawing
  • US20220291898A1 patent drawing

AI summary

According to one embodiment, an arithmetic device includes an arithmetic element part, and a controller. The arithmetic element part includes first and second elements. The first element includes a first conductive member and a first stacked body. The first conductive member includes first to third portions. The first stacked body includes a first magnetic layer, and a first counter magnetic layer. The second element includes a second conductive member and a second stacked body. The second conductive member includes fourth and fifth portions, and a sixth portion between the fourth and fifth portions. The second stacked body includes a second magnetic layer, and a second counter magnetic layer. The controller is configured to perform an XNOR operation of first and second inputs. The first input corresponds to electrical resistances of the stacked bodies. The second input corresponds to potentials of the magnetic layers.