Concept for Parameter Design and Optimization of Surface Acoustic Wave Devices

Introduction. Bandpass filters on surface acoustic waves (SAW) are important components in frequency selection devices for receiving and transmitting electronic equipment. Among the variety of existing requirements for SAW filters, particular attention is paid to those for the quality of characteristics, including reduction of insertion losses, unevenness in the bandwidth, etc., and those for reducing the development time of the filter. Reduction in the development time along with a simultaneous reduction in expenses can be achieved through the application of preliminary computer simulation. This task can be implemented only through an integrated approach to developing a computer-aided design system capable of solving optimization problems with a large number of degrees of freedom.Aim. Development of a prototype of a fully functional complex for simulating and calculating SAW filters. Approbation of work on the creation of actual frequency filters on the example of resonator SAW filters with low losses on leaky SAWs. Development of a methodological approach to creation of a technology of computer-aided design of SAW filters using optimization algorithms.Materials and methods. The theoretical part of the work was carried out using statistical analysis methods, circuit theory, and coupled mode theory. In the course of the work, mathematical processing and calculation were carried out in the MATLAB environment.Results. A concept for creating a system of automated design of SAW filters with specified frequency characteristics was developed. An original method is proposed for optimizing the topology of SAW filter elements to obtain the maximum achievable frequency responses. The proposed approach was tested when calculating a resonator filter on leaky SAWs at 64° YX-cut lithium niobate.Conclusion. The proposed integrated approach to designing SAW filters allows a rapid and relatively accurate prediction of filter characteristics at the modeling stage, which has a significant advantage compared to conducting numerous real experimental studies or numerical studies in the presence of a large number of degrees of freedom.