Resistance-Variable Memory Cells: Snapback Pulses for Multi-State Programming
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing memory technologies are limited to programming resistance variable memory cells in two states, which restricts their application in complex operations and increases the bit requirement for encoding data, leading to inefficiencies in memory density, cost, and performance.
Innovation Solution
Implement multi-state programming of memory cells by applying voltage pulses to induce snapback events, followed by additional voltage pulses to adjust threshold voltages, enabling cells to be programmed to multiple data states.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If two-state programming is used for resistance variable memory cells, then the programming process is simple, but memory density is limited and more bits are required for encoding data
Solution Approach 1:
The patent applies parameter changes by modifying the voltage pulse characteristics (amplitude, duration, polarity) to program memory cells to multiple resistance states. By varying these electrical parameters, the memory cell can be programmed to different data states (e.g., 0, 1, 2, 3) instead of just two states, thereby increasing memory density without proportionally increasing device complexity
Solution Approach 2:
The patent employs periodic action through the application of sequential voltage pulses to program memory cells to multiple states. Multiple pulses with different characteristics are applied in sequence to achieve different resistance levels, enabling multi-state programming that increases storage capacity while maintaining a systematic programming approach
2Ease of operation
If two-state programming is used for resistance variable memory cells, then the encoding process is straightforward, but more bits are needed to represent data
Solution Approach 1:
By changing the resistance state parameters of memory cells to multiple discrete levels, the patent enables more data to be stored per cell. For example, a memory cell with 4 distinguishable resistance states can store 2 bits of information (representing values 0, 1, 2, or 3) instead of just 1 bit, thereby reducing the total number of bits required for data encoding while maintaining operational simplicity
3Quantity of substance
If multi-state programming is implemented, then memory density and performance improve, but the programming process becomes more complex
Solution Approach 1:
The patent manages the complexity of multi-state programming by systematically varying voltage pulse parameters (amplitude, duration, polarity) to achieve different resistance states. This parameter-based approach provides a structured method for programming multiple states, making the process more controllable and less complex than it would otherwise be
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
Enhances memory density, reduces cost, and improves performance by allowing cells to be programmed to multiple states, supporting complex operations like machine learning and reducing the number of bits needed for encoding.
Implementation Method 1
it is determined the memory cell snaps back in response to the applied voltage pulse
Data Source
AI summary
The present disclosure includes apparatuses, methods, and systems for multi-state programming of memory cells. An embodiment includes a memory having a plurality of memory cells, and circuitry configured to program a memory cell of the plurality of memory cells to one of a plurality of possible data states by applying a voltage pulse to the memory cell, determining the memory cell snaps back in response to the applied voltage pulse, turning off a current to the memory cell upon determining the memory cell snaps back, and applying a number of additional voltage pulses to the memory cell after turning off the current to the memory cell.