IC Placement System Using Genetic Algorithm for Chip Area and Heat Reduction

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional automatic placement systems for IC design face issues with large chip area, excessive heat generation, noise interference, and prolonged computation times, limiting their effectiveness in achieving optimal circuit component placement.

Innovation Solution

An automatic placement system utilizing a genetic algorithm with input and processing modules that generate and evolve chromosome sequences based on component data and parameters, including crossover and mutation rates, to determine the best placement method, incorporating a revised tree-structure methodology and roulette wheel algorithm to optimize chip area, wire length, heat management, and component switching.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Extent of automation

If conventional automatic placement systems use computed placement methods, then the placement can be automated, but the chip area becomes too large

Engineering Contradiction:

Improveautomatic placementVSAvoidchip area

Core Design Contradiction:

Extent of automationVSArea of stationary object

Solution Approach 1:

The patent uses genetic algorithms to optimize placement parameters such as component positions, orientations, and spacing. By iteratively evolving chromosome sequences representing different placement configurations, the system finds optimal arrangements that minimize chip area while maintaining all necessary electrical connections and design constraints.

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The placement system dynamically adjusts component positions and orientations through the genetic evolution process. Instead of static computed placement, the system continuously evolves placement solutions across multiple generations, adapting the configuration to achieve better area utilization and performance metrics.

Inventive Principle:

Principle #15Dynamics

2Extent of automation

If conventional automatic placement systems use computed placement methods, then the placement can be automated, but the circuit component generates too much heat

Engineering Contradiction:

Improveautomatic placementVSAvoidheat generation

Core Design Contradiction:

Extent of automationVSTemperature

Solution Approach 1:

The genetic algorithm optimizes thermal parameters by evaluating heat generation metrics during the fitness assessment of each chromosome sequence. The system adjusts component positions and orientations to distribute heat-generating elements more effectively, reducing hot spots and improving thermal management through evolved placement configurations.

Inventive Principle:

Principle #35Parameter changes

3Extent of automation

If conventional automatic placement systems use computed placement methods, then the placement can be automated, but noise interference increases

Engineering Contradiction:

Improveautomatic placementVSAvoidnoise interference

Core Design Contradiction:

Extent of automationVSObject-affected harmful factors

Solution Approach 1:

The system optimizes placement parameters to minimize noise interference by evaluating electromagnetic compatibility metrics during genetic evolution. The algorithm adjusts component positions and orientations to reduce coupling between sensitive and noisy elements, improving signal integrity through evolved configurations that account for noise characteristics.

Inventive Principle:

Principle #35Parameter changes

4Extent of automation

If conventional automatic placement systems use computed placement methods, then the placement can be automated, but the computation time becomes too long

Engineering Contradiction:

Improveautomatic placementVSAvoidcomputation time

Core Design Contradiction:

Extent of automationVSLoss of time

Solution Approach 1:

The genetic algorithm uses periodic generations with structured evolution cycles, performing crossover and mutation operations at regular intervals. This periodic structure allows the system to explore the solution space efficiently while maintaining computational tractability, balancing exploration and exploitation across generations.

Inventive Principle:

Principle #19Periodic action

Solution Approach 2:

The system applies partial genetic operations (crossover and mutation) to selected chromosome sequences rather than processing all possible configurations. By applying genetic operators selectively to promising solutions and using termination conditions to limit evolution depth, the system achieves good placement results without exhaustive computation.

Inventive Principle:

Principle #16Partial or excessive action

Data Source

PatentUS8856716B1Automatic placement system of IC design and method thereof

Publication Date: 2014.10.07 NAT TAIWAN UNIV

AI summary

An automatic placement system of IC design and a method thereof is provided. The automatic placement system of IC design concerns the chip area utility ratio, the input-output relationship between components, the power consumption produced from thermal noise of circuits and the MOS-type transformation ratio, and performs the genetic algorithm for providing an optimal solution to the placement problem. Herewith the effect of optimizing the placement according to the data of components and parameter is achieved.