Automated Edge Movement for Circuit Device Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for optimizing transistor performance in integrated circuit design often fail to balance individual device performance with the collective performance of surrounding devices, leading to inefficiencies in circuit design where some devices need to be stronger or weaker than a reference device.
Innovation Solution
A method that allows for the adjustment of edge positions of semiconductor device features, such as stressed liners, in specific directions to improve both individual device performance and matching with other devices, ensuring that performance goals are met while adhering to design rules and tolerances.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the boundaries of n-wells are moved outwards to maximize individual transistor performance, then individual device performance is improved, but the ability to achieve desired matching between devices is compromised
Solution Approach 1:
The patent applies local quality by enabling different edge movement directions for different devices based on their specific matching requirements. Each device can have its n-well boundaries moved in opposite directions relative to its channel, allowing individualized performance optimization while maintaining proper matching between connected devices. This resolves the contradiction by making the optimization adaptive to local circuit requirements rather than applying a uniform approach.
2Productivity
If automated optimization is applied to all devices, then overall circuit performance improves, but the complexity of managing direction-specific adjustments increases
Solution Approach 1:
The patent implements dynamics by making the optimization process adaptive and responsive to circuit requirements. The automated system dynamically determines the optimal edge movement direction for each device based on circuit topology and performance goals, rather than following static predetermined rules. This allows the system to manage complexity through intelligent automation that adapts to different circuit configurations automatically.
Solution Approach 2:
The patent employs feedback mechanisms where the optimization process evaluates circuit performance and adjusts edge positions iteratively. The system monitors whether moving device edges in specific directions achieves the desired matching and performance goals, then uses this information to guide further adjustments. This feedback loop enables automated optimization to effectively manage complexity by learning from results and making data-driven decisions.
Data Source
AI summary
A method of improving a circuit design for a very large scale integrated circuit is provided which represents a plurality of semiconductor devices interconnected in a circuit. It is determined whether an edge of a feature of one of the plurality of semiconductor devices in the design can be moved in a first direction by a distance within a permitted range, such that a performance goal and a matching goal for the circuit are served. If so, the edge is moved in the first direction by the distance calculated to best serve the performance goal and the matching goal. The foregoing steps may be repeated for each of the plurality of semiconductor devices. If necessary, the foregoing steps may be repeated until the performance goal and matching goal for the circuit are deemed to be adequately served.