Application of metaheuristics through MATLAB optimisation toolboxes for the design of coupled resonator filters

There is a large number of applications in the design of telecommunication devices where it is necessary to apply optimisation techniques to find the best values for a set of parameters [2, 5]. Normally, it is not possible to solve those optimisation problems with analytical techniques, and the alternative is to use metaheuristic methods [6]. In this work we tackle the problem of designing coupled resonator filters [1], which are used in the RF-frontends of communication devices. There are different phases in the design of these filters. In a first approximation, the desired topology of the filter and the basic characteristics of the technology to use are established, and then it is necessary to obtain the theoretical coupling matrix which implements the desired transfer function. We analyse the application of metaheuristics to obtain that matrix. Furthermore, the numerical software Matlab is frequently used to work on the design of those devices, and a good option would be to also develop in Matlab the optimisation tools. Thus, this work analyses the application of Matlab optimisation toolboxes to tackle this optimisation problem with different metaheuristics. A detailed experimental analysis is carried out to obtain the tools in Matlab which best fit the problem, and some tools are personalised to improve the quality of the design.