Optimization in Electronic Design Automation

Higher education teachers: Tuma Tadej
Collaborators: Bürmen Arpad
Credits: 5
Subject code: 64815



Subject description

Prerequisits:

  • Basic knowledge of EDA theory. Some experience in using SPICE circuit simulators.

Content (Syllabus outline):

  • Term definition and optimization method overview. Unconstrained methods: first and second order gradient algorithms, direct algorithms and genetic algorithms. Constrained methods: parameter constrains, penalty functions, parameter space transformations. Analog circuit sensitivity in frequency and time.
  • Introduction to an analog circuit optimization tool. The simplex algorithm and its parallelization. Choosing a seed. Measurement definition and cost function formulation. Including design corners into the cost function. Cost function properties (cost profile and noise). A demo run of the tool and result interpretation.
  • A medium size circuit is run by each student on a personal computer. A large analog optimization case is run on a computer farm.

Objectives and competences:

Theoretical knowledge of optimization procedures in EDA. Employing optimization techniques in practical circuit simulation.

Intended learning outcomes:

Student is capable of independent analogue integrated circuit design.

Learning and teaching methods:

  • Individual consultation,
  • directed seminar work,
  • independent project work.





Study materials

  1. TUMA, Tadej, BÜRMEN, ArpadCircuit Simulation with SPICE OPUS, Theory and Practice. Springer, 2009, Approx. 480 p. 158 illus., Hardcover ISBN: 978-0-8176-4866-4, A Birkhäuser book.
  2. Web page of SPICE OPUS (www.spiceopus.si)