Netlist Proximity-Effect Parameters for Semiconductor Simulation Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current semiconductor manufacturing processes face challenges in accurately representing proximity effects in layout diagrams, leading to inadequate simulations due to the coarse granularity of existing methods, which fail to account for the specific influences of neighboring cells on a subject cell.
Innovation Solution
The introduction of neighbor-specific proximity-effect (NSPE) parameters in sidefiles and parameterized netlists, which provide a more granular representation of inter-cell proximity effects by detailing the spatial relationships between the subject cell and its neighbors, including specific transistor configurations and distances, enhances the accuracy of simulations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If coarse granularity methods are used to represent proximity effects, then the complexity of the netlist is reduced, but the measurement precision of proximity effects deteriorates
Solution Approach 1:
The patent segments the proximity effect representation by introducing neighbor-specific proximity-effect (NSPE) parameters that divide the overall proximity effect into individual contributions from each neighboring cell. This allows the netlist to capture detailed spatial relationships between the subject cell and each neighbor separately, improving measurement precision without excessively increasing overall complexity through systematic parameter organization.
Solution Approach 2:
The patent applies local quality by assigning specific NSPE parameters to different neighboring cells based on their individual spatial relationships with the subject cell. Each neighbor's proximity effect is characterized by its own parameters (distance, orientation, cell type), allowing the representation to be locally optimized for each neighbor while maintaining overall netlist manageability through localized parameter assignments.
2Reliability
If detailed spatial relationships between cells are captured, then the reliability of simulation results is improved, but the device complexity of the netlist increases
Solution Approach 1:
The patent segments the simulation accuracy improvement by introducing NSPE parameters that separately capture the proximity effect from each neighboring cell. This segmentation allows the netlist to systematically represent detailed spatial relationships (distance, orientation, neighbor cell characteristics) without creating an unmanageable monolithic structure, as each neighbor's contribution is independently parameterized and organized.
Solution Approach 2:
The patent applies universality by designing the NSPE parameter structure to handle multiple types of spatial relationships and neighbor cell configurations through a unified parameter framework. The same NSPE parameter structure can represent various neighbor types (adjacent cells, diagonal cells, different cell densities), reducing netlist complexity by avoiding the need for separate specialized structures for each scenario.
3Measurement precision
If neighbor-specific proximity-effect parameters are introduced, then the measurement precision of inter-cell proximity effects is improved, but the difficulty of detecting and measuring these effects increases
Solution Approach 1:
The patent applies preliminary action by pre-defining the NSPE parameter structure and categories before detailed measurement and extraction. The framework预先 establishes the types of spatial relationships to measure (distance, orientation, neighbor cell type), which guides the subsequent detection process and reduces the complexity of measuring each individual parameter by providing a structured approach ahead of time.
Data Source
AI summary
A method (of manufacturing a semiconductor device, a corresponding layout diagram including cells such that, for a subset of the cells, each subject one of the cells (subject cell) in the subset has a neighborhood including first and second neighbor cells on corresponding first and second sides of the subject cell relative to the first direction) includes: for each subject cell in the subset, generating a sidefile which represents neighborhood-specific proximity-effect information; and, for each cell in the subset of the cells, the generating a sidefile including: populating the sidefile with a first neighbor-specific proximity-effect (NSPE) parameter (corresponding to an inter-cell proximity-effect induced by the first neighbor cell) identifying a nearest first transistor of the first neighbor cell; and populating the sidefile with a second NSPE parameter (corresponding to an inter-cell proximity-effect induced by the second neighbor cell) identifying a nearest first transistor of the second neighbor cell.