Triple Row Cell Architecture for IC Area Performance Tradeoffs

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing integrated circuit design faces tradeoffs between area usage and performance, with single row transistor cells consuming less area but operating slower, and double row cells operating faster but at the cost of increased space and efficiency losses.

Innovation Solution

A triple row transistor cell layout is introduced, where a PMOS and NMOS transistor each span across one row boundary, with spare areas that can accommodate additional components, improving speed performance and area efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Area of stationary object

If single row transistor cells are used, then area usage is reduced, but speed performance deteriorates

Engineering Contradiction:
Improvecell areaVSAvoidtransistor switching speed
Core Design Contradiction:
Area of stationary objectVSSpeed

Solution Approach 1:

The patent transitions from single-row to triple-row cell architecture, utilizing the vertical dimension (row spanning) to improve transistor performance. By allowing transistors to span across multiple rows with shared source/drain regions, the design achieves better speed performance while maintaining area efficiency through the multi-row configuration.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Speed

If double row transistor cells are used, then speed performance is improved, but area usage increases and efficiency is lost

Engineering Contradiction:
Improvetransistor switching speedVSAvoidcell area
Core Design Contradiction:
SpeedVSArea of stationary object

Solution Approach 1:

The patent merges transistor structures across three rows by sharing source and drain regions between adjacent rows. This consolidation allows the cell to achieve the speed benefits of multi-row architecture while reducing the total area compared to traditional double-row cells, as the shared regions eliminate redundant structures.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The source and drain regions serve multiple functions by being shared across adjacent rows. Each source/drain region acts as both the source for one transistor and the drain for the adjacent transistor, maximizing the utility of each fabricated region and improving overall area efficiency.

Inventive Principle:
Principle #6Universality (Multi-functionality)

3Speed

If triple row transistor cells are used, then speed performance is improved, but area usage may increase

Engineering Contradiction:
Improvetransistor switching speedVSAvoidcell area
Core Design Contradiction:
SpeedVSArea of stationary object

Solution Approach 1:

The triple-row cell architecture segments the cell boundary across three distinct rows, with each row contributing to the overall transistor function. This segmentation allows for optimized placement of transistors and interconnects within each row while maintaining compact overall cell dimensions through shared source/drain regions.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentUS7564077B2Performance and area scalable cell architecture technology
Publication Date: 2009.07.21 TEXAS INSTRUMENTS INC
  • US7564077B2 patent drawing
  • US7564077B2 patent drawing
  • US7564077B2 patent drawing

AI summary

An integrated circuit. The integrated circuit comprises an area having a layout aligned in rows. Each row is definable by a pair of row boundaries. The integrated circuit also comprises a plurality of cells, comprising a first set of cells. Each cell in the first set of cells spans at least two rows and comprises a PMOS transistor having a source/drain region that spans across one of the row boundaries and an NMOS transistor having a source/drain region that spans across one of the row boundaries.