Automatic tuning algorithms and statistical circuit design

First three generalized tuning algorithms are comparatively reviewed on the basis of their architecture, computational complexity, and effectiveness. Also, a method is presented for the tuning resistor and frequency selection problem, a problem relevant to all three methods. Secondly, Monte Carlo techniques for yield estimation and yield maximization of circuits are investigated. A quadratic approximation model is set up for the circuit and is used to achieve variance reduction in yield estimation and to provide an extrapolated yield approximation technique.