Crossbar RRAM Programming With RTN Feedback Suppression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies fail to effectively address the issue of stochastic fluctuations in crossbar circuits, specifically in suppressing random telegraph noise in crossbar circuits, which leads to errors and instability in crossbar circuits.
Innovation Solution
Implementing a method for suppressing random telegraph noise (RTN) in crossbar circuits by programming RRAM devices to a target conductance value, determining RTN values, and applying noise-reduction voltages until the RTN values fall within an acceptable range, using a processing device to generate and apply these voltages.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If programming voltage is applied to program RRAM device to target conductance value, then conductance precision is improved, but random telegraph noise increases
Solution Approach 1:
The patent applies preliminary screening actions by measuring RTN values after initial programming and identifying devices that exceed noise thresholds. This preliminary detection enables selective re-programming of only those devices requiring noise suppression, rather than re-programming all devices, thus resolving the contradiction between achieving low noise and maintaining programming efficiency.
Solution Approach 2:
The patent implements a feedback mechanism where RTN values are measured after programming, and this measurement feedback triggers conditional re-programming actions. The system continuously monitors noise levels and applies additional programming voltages only when noise exceeds thresholds, creating a closed-loop control system that balances conductance precision with noise suppression.
2Object-generated harmful factors
If additional programming processes are applied to reduce RTN, then noise level is improved, but processing time increases
Solution Approach 1:
The patent performs preliminary RTN measurement after initial programming to identify which devices actually require noise reduction. By screening devices beforehand and applying re-programming only to those exceeding noise thresholds, the system avoids unnecessary processing time for devices that already meet noise requirements, thus resolving the time-noise tradeoff.
Solution Approach 2:
The patent applies partial action by implementing selective re-programming only for a subset of devices that exhibit excessive RTN, rather than applying additional programming to all devices. This partial approach reduces overall processing time while still achieving noise suppression where needed, balancing the contradiction between noise reduction and time efficiency.
3Stability of the object's composition
If noise-reduction voltage is applied below threshold, then device stability is improved, but programming efficiency decreases
Solution Approach 1:
The patent uses feedback from RTN measurements to determine whether noise-reduction voltages are needed. By continuously monitoring noise levels and comparing them against thresholds, the system intelligently applies stability-enhancing voltages only when necessary, preventing unnecessary operations that would reduce programming efficiency while maintaining device stability where required.
Solution Approach 2:
The patent performs preliminary assessment of device noise characteristics before applying stability-enhancing voltages. This preliminary evaluation enables the system to apply noise-reduction voltages only to devices that benefit from them, avoiding unnecessary applications that would degrade programming efficiency while still improving stability for affected devices.
Data Source
AI summary
The present disclosure provides mechanisms for reducing and suppressing random telegraph noise (RTN) for a crossbar circuit. A processing device may perform a programming process to program the conductance of a resistive random-access memory (RRAM) device in the crossbar circuit to a target conductance value. The processing device may then determine whether a random telegraph noise (RTN) value associated with the RRAM device is within a predetermined range of acceptable RTN values. If the RTN value associated with the RRAM device is not within a predetermined range of acceptable RTN values, one or more noise-reduction voltages may be applied to the RRAM device until the RTN value associated with the RRAM device is within the predetermined range of acceptable RTN values.