Implicit Space Mapping for Variable-Fidelity EM-Driven Design of Compact Circuits

Abstract

Space mapping (SM) belongs to the most successful surrogate-based optimization (SBO) methods in microwave engineering. Among available SM variations, implicit SM (ISM) is particularly attractive due to its simplicity and separation of extractable surrogate model parameters and design variables of the circuit/system at hand. Unlike other SM approaches, ISM exploits a set of preassigned parameters to align the surrogate with the high-fidelity EM model. However, application of ISM is challenging if equivalent network model of the structure is unavailable or of poor quality. In this letter, a modified ISM that exploits variable-fidelity EM simulation models is proposed. Here, preassigned parameters are introduced in the coarsely discretized EM surrogate by dividing the substrate below selected components of the structure into segments with different permittivity and common thickness. The proposed method has been verified (also experimentally) using a rat-race coupler and favorably compared with state-of-the-art SBO methods.