Nonvolatile Memory True Random Number Generation via Threshold Voltage Manipulation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing memory systems face challenges in generating high randomness for encryption processes, as pseudo-random numbers may not satisfy the required randomness, leading to increased costs when dedicated hardware random number generators are used to produce true random numbers.
Innovation Solution
The memory system utilizes the nonvolatile memory itself as a natural random number generator by manipulating threshold voltage distributions across specific read levels, leveraging natural phenomena like data retention and read disturbance to generate true random numbers without additional hardware.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If pseudo-random numbers are used for encryption processes, then system cost is reduced, but randomness quality deteriorates and cannot satisfy required encryption security
Solution Approach 1:
The nonvolatile memory serves dual purposes: as storage medium for data and as a random number generator for encryption processes. By utilizing the inherent physical characteristics of the memory cells (threshold voltage variations) for both data storage and random number generation, the system eliminates the need for separate hardware random number generators, thereby reducing system cost while maintaining high randomness quality required for encryption security.
Solution Approach 2:
The memory system generates its own random numbers using the natural physical variations in its own memory cells, without requiring external random number generation hardware. The threshold voltage variations that naturally exist in the memory cells are directly exploited to produce high-quality random numbers, making the system self-sufficient for cryptographic requirements.
2Reliability
If dedicated hardware random number generators are used to produce true random numbers, then randomness quality is improved, but system cost increases
Solution Approach 1:
The nonvolatile memory serves dual purposes: as storage medium for data and as a random number generator for encryption processes. By utilizing the inherent physical characteristics of the memory cells (threshold voltage variations) for both data storage and random number generation, the system eliminates the need for separate hardware random number generators, thereby reducing system cost while maintaining high randomness quality required for encryption security.
Solution Approach 2:
The memory system generates its own random numbers using the natural physical variations in its own memory cells, without requiring external random number generation hardware. The threshold voltage variations that naturally exist in the memory cells are directly exploited to produce high-quality random numbers, making the system self-sufficient for cryptographic requirements.
3Reliability
If threshold voltage distribution is manipulated to generate random numbers, then randomness is improved, but manufacturing precision requirements increase
Solution Approach 1:
The invention converts the natural manufacturing variations and threshold voltage distributions, which are typically considered imperfections or sources of error in memory cells, into a beneficial resource for generating high-quality random numbers. Instead of requiring tight control over threshold voltages, the system exploits the inherent variations to produce unpredictable random values, thereby improving randomness without imposing stricter manufacturing precision requirements.
Solution Approach 2:
The system changes the approach from controlling threshold voltage to a fixed value to measuring and utilizing threshold voltage variations as a parameter for random number generation. By reading the threshold voltages of memory cells and exploiting their natural distribution, the system transforms a potential manufacturing challenge into a source of high-entropy random numbers.
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 reduces system costs while improving randomness, enabling the generation of high-entropy random numbers suitable for encryption processes without the need for dedicated hardware, thus enhancing the reliability and efficiency of the memory system.
Implementation Method 1
leveraging natural phenomena like data retention and read disturbance to generate true random numbers
Implementation Method 2
leveraging natural phenomena like data retention and read disturbance to generate true random numbers
Data Source
AI summary
A memory system includes a nonvolatile memory and a controller that performs first, second, and third processes on memory cells of the nonvolatile memory. The first process is performed on first memory cells to store a first value therein, such that a highest threshold voltage among the threshold voltages of the first memory cells is set as a first threshold voltage. The second process is performed on second memory cells to store a second value therein, such that a lowest threshold voltage among the threshold voltages of the second memory cells is set as a second threshold voltage higher than the first threshold voltage. The third process performed on third memory cells such that a lowest threshold voltage in the third memory cells is lower than the first threshold voltage, and a highest threshold voltage in the third memory cells is higher than the second threshold voltage.