LEF Diagram Processing via Segment Merging

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The existing method of processing LEF diagrams in semiconductor integrated circuit manufacturing results in increased data size, leading to inefficiencies in data transmission due to the segmentation of polygonal structures into multiple rectangles.

Innovation Solution

A method that breaks down the LEF diagram into rectangular segments, numbers them, and combines adjacent segments to form larger segments, reducing the total number of segments and thereby reducing the data size of the LEF file, using EDA tools for segmentation and numbering.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If the LEF diagram is segmented into multiple rectangular segments according to vertex coordinates, then the representation accuracy of polygonal structures is improved, but the data size of the LEF file increases

Engineering Contradiction:
Improverepresentation accuracyVSAvoiddata size
Core Design Contradiction:
Manufacturing precisionVSQuantity of substance

Solution Approach 1:

The patent merges adjacent rectangular segments that share common edges to form larger rectangular segments. This reduces the total number of segments needed to represent the same polygonal structure, thereby decreasing data size while maintaining representation accuracy. The method identifies adjacent segments through coordinate comparison and combines them by creating parent-child relationships between segment nodes.

Inventive Principle:
Principle #5Merging (Combining)

2Manufacturing precision

If more rectangular segments are used to represent polygonal structures, then the geometric precision is improved, but the data transmission efficiency deteriorates

Engineering Contradiction:
Improvegeometric precisionVSAvoiddata transmission efficiency
Core Design Contradiction:
Manufacturing precisionVSProductivity

Solution Approach 1:

By combining adjacent rectangular segments into larger segments, the patent reduces the total segment count, which directly decreases data transmission volume and improves efficiency. The merging process maintains geometric precision by preserving the exact boundary coordinates of the original segments while reducing redundancy.

Inventive Principle:
Principle #5Merging (Combining)

3Manufacturing precision

If the LEF diagram is divided into many small rectangular segments, then the detail representation is improved, but the file processing complexity increases

Engineering Contradiction:
Improvedetail representationVSAvoidfile processing complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent reduces file processing complexity by merging small adjacent segments into larger segments. This decreases the number of objects that need to be processed, stored, and transmitted, while the hierarchical parent-child relationship structure maintains the ability to represent detailed geometric information when needed.

Inventive Principle:
Principle #5Merging (Combining)

Data Source

PatentUS10410356B2Method for processing the LEF diagram of a layout
Publication Date: 2019.09.10 SHANGHAI HUAHONG GRACE SEMICON MFG CORP
  • US10410356B2 patent drawing
  • US10410356B2 patent drawing
  • US10410356B2 patent drawing

AI summary

A method for processing the Library Exchange Format (LEF) diagram of a layout includes the following steps: Step 1, breaking the LEF diagram into multiple rectangular segments; Step 2, numbering all rectangular segments; and Step 3, combining rectangular segments to obtain a larger rectangular segment and replacing corresponding uncombined rectangular segments with the larger combined rectangular segment. The method reduces the data size of the LEF file and increase the data transmission efficiency of the LEF file.