Multi-hard-objective optimum design of surface acoustic wave duplexers using adaptive differential evolution

This paper proposes a novel computer-aided design method of Surface Acoustic Wave (SAW) duplexers which are used widely in the modern Radio Frequency (RF) circuits of various mobile communication systems. The desirable frequency response characteristics of a SAW duplexer are specified by a number of necessary conditions. Besides, the frequency response characteristics of the SAW duplexer depends on its geometrical structure. The major advantage of the proposed design method is that it provides the designer with a set of optimal solutions, or structures, instead of a single solution. Therefore, the designer can choose a solution which of them best fits the application under consideration. According to the proposed design method, the structural design of a SAW duplexer is described as a Multi-Hard-objective Optimization Problem (MHOP) in which each of multiple objective functions is restricted to be less than its upper bound. Then, in order to obtain a number of non-dominated feasible solutions for MHOP, a new evolutionary algorithm based on Differential Evolution (DE) is employed. Finally, through analysis of the non-dominated feasible solutions, or structures, obtained by the DE-based algorithm, suitable structures of the SAW duplexer are considered from various viewpoints.