Chalcogenide Threshold Voltage Random Number Generation in Memory

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing methods for generating random numbers in memory devices often incur high resource consumption and latency, particularly when relying on software algorithms, which can be inefficient and resource-intensive.

Innovation Solution

A memory device utilizes a chalcogenide element with a stochastic threshold voltage to generate random numbers by applying a voltage until the threshold is reached, employing an oscillating signal to detect the exact moment of conductivity change, thereby reducing resource allocation and latency through hardware-based generation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If software algorithms are used to generate random numbers, then random number generation can be implemented, but resource consumption and latency increase

Engineering Contradiction:

Improverandom number generation capabilityVSAvoidprocessing bandwidth consumption

Core Design Contradiction:

ReliabilityVSUse of energy by moving object

Solution Approach 1:

The patent replaces software-based random number generation algorithms with a hardware-based solution using chalcogenide elements. The stochastic threshold voltage characteristics of these elements provide inherent randomness, eliminating the need for computationally intensive software algorithms and reducing processing bandwidth consumption.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The chalcogenide elements generate random numbers through their intrinsic physical properties - specifically the stochastic variation in threshold voltage. This self-service mechanism requires minimal external processing, as the randomness emerges naturally from the material characteristics during normal operation.

Inventive Principle:

Principle #25Self-service

2Reliability

If software algorithms are used to generate random numbers, then random number generation can be implemented, but latency increases

Engineering Contradiction:

Improverandom number generation capabilityVSAvoidgeneration latency

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The chalcogenide elements are pre-configured with stochastic threshold voltage characteristics that inherently produce randomness. This preliminary preparation of the hardware component eliminates the need for time-consuming software computation during actual random number generation operations.

Inventive Principle:

Principle #10Preliminary action

3Adaptability or versatility

If additional programming or preconditioning is applied to memory devices for random number generation, then functionality is enhanced, but device complexity increases

Engineering Contradiction:

Improverandom number generation functionalityVSAvoidprogramming and preconditioning requirements

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The chalcogenide elements serve dual purposes: they function as memory storage elements and simultaneously provide random number generation capability through their stochastic threshold voltage. This multi-functionality eliminates the need for separate programming or preconditioning steps, reducing device complexity while maintaining versatility.

Inventive Principle:

Principle #6Universality (Multi-functionality)

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 approach effectively reduces resource consumption and latency in generating random numbers by leveraging the inherent randomness of chalcogenide elements already integrated into memory architectures, without the need for additional programming or preconditioning.

Implementation Method 1

A memory device may include a chalcogenide element of a memory system. The chalcogenide element may have a stochastic threshold voltage

Methodology Applied

Scientific EffectStochastic threshold voltage:

Implementation Method 2

increasing the applied voltage at least until the applied voltage satisfies a threshold voltage associated with the chalcogenide element

Methodology Applied

Scientific EffectThreshold voltage satisfaction:

Data Source

PatentUS20240057489A1Random number generation based on threshold voltage randomness

Publication Date: 2024.02.15 MICRON TECHNOLOGY INC

AI summary

Methods, systems, and devices for random number generation based on threshold voltage randomness are described. For example, a memory device may apply a voltage to a chalcogenide element and increase the applied voltage at least until the applied voltage satisfies a threshold voltage associated with the chalcogenide element. The memory device may detect the state of an oscillating signal at a time at which the applied voltage satisfies the threshold voltage, and the memory device may output a logic value corresponding to the state of the oscillating signal. The threshold voltage of the chalcogenide element may vary in a statistically random manner across voltage applications, and hence the state of the oscillating signal at the time an applied voltage reaches the threshold voltage may likewise vary in a statistically random manner, and thus the corresponding logic value that is output may be a random value suitable for random number generation.