On the design and optimization of waveguides using evolutionary approaches

This work presents a new and efficient CAD-oriented algorithm for the design and optimization of high frequency coupled coplanar waveguides (CCPW). The technique is based on genetic algorithms to obtain the global optimal solution of the problem. The proposed algorithm optimizes a multi-objective highly nonlinear problem having multiple local minima with one constraint. The new approach obtains the optimal structure dimensions that minimize the attenuation and at the same time is as close as possible to the circuit matching condition. After validation, the proposed technique is compared to a global search optimizer and then successfully applied to the design of practical monolithic implementations.