University of Sussex
Browse

File(s) not publicly available

Electronic circuit design

standard
posted on 2023-06-07, 20:46 authored by Adrian Dominic Thompson
A system for optimising electronic circuits to be designed has two parts or phases, a training phase 10 in which optimisation scenarios for selected electronic circuits are derived using an evolutionary algorithm, and an operation phase 11, in which the derived optimisation scenarios are used to optimise new electronic circuits to be designed. The training phase 10 uses an evolutionary algorithm to produce a specialised and relatively fast optimisation scenario for each of a plurality of input circuits or groups of circuits (step 13). One or more of the evolved specialist optimisation scenarios are then selected to form a suite of optimisation scenarios (step 15) for use to optimise new circuits to be designed. In the operation phase 11, a new circuit to be optimised is input at step 18, and a plurality of optimisation scenarios from the suite of optimisation scenarios 15 is then used to try to optimise the new circuit (step 19). The best optimisation result is taken as the optimisation for the circuit (step 20).

History

Publication status

  • Published

Publisher

World Intellectual Property Organization, Patent Cooperation Treaty

Department affiliated with

  • Informatics Publications

Notes

Originality: Describes the first generally effective method by which industry-standard tools for electronics design automation can be improved using evolutionary algorithms to tune the scripts that control them. Builds on an prior proposal that was restricted in scope so impractical. Rigour: Exploits the surprising discovery that a suite of specialised solutions can readily be evolved that collectively gives excellent general performance: data gathered from the prototype system working on standard benchmarks shows a significant improvement over previously published work. Significance: Our ongoing plans to exploit this patent should give direct cost/quality benefits to the optimisation tool vendors and users, both in massively lucrative industries. Outlet/citations: Published with international search report, no conflicting prior art found. Journal papers have been held back until the patent application was published; "Improved logic optimisation using a suite of diverse specialist optimisation scripts produced by evolutionary algorithms" is under submission to IEEE Trans. CAD.

Full text available

  • No

Peer reviewed?

  • No

Legacy Posted Date

2012-02-06

Usage metrics

    University of Sussex (Publications)

    Categories

    No categories selected

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC