ABSTRACT
In many integrated circuit (IC) design situations, a designer must make complex trade-offs between conflicting behavioral requirements, dealing with functions that are often nonlinear. The number of parameters involved in the design process may be large, necessitating the use of algorithms that provide qualitatively good solutions in a computationally efficient manner. This chapter illustrates the theory and utilization of optimization algorithms in computer-aided design (CAD) of ICs. The form of a general nonlinear optimization problem is first presented, and some of the commonly used methods for optimization are overviewed. Case studies on the following specific design problems are examined: transistor sizing and analog design centering. Most problems in IC design involve the minimization or maximization of a cost function subject to certain constraints. The chapter also presents a few prominent techniques for constrained optimization: Lagrange multiplier methods, penalty function methods, and the method of feasible directions.
