Tabu Search for Estimating Optimal Values of Transduction and Reflection Weighing Functions Of SAW Filter

Surface Acoustic Wave (SAW) filters are used in communication devices due to its small size and relatively superior performance at high frequencies in comparison to other competing filters. However, the design of SAW filter is always a critical issue due to requirement of small size as well as high performance parameters. Single Phase Unidirectional Transducers (SPUDTs) are the key element in design of SAW filters. The most basic challenge in designing with SPUDTs is to estimate both transduction and reflection weighting functions simultaneously that lead to desired frequency response of a SAW filter. As there is internal reflection to the transducer to cancel the effects of regeneration reflections, so the design of SPUDT is a very challenging task. This paper presents a typical local search technique, i.e., Tabu Search for estimation of sequence of transduction weighting coefficients (t) and reflection weighting coefficients (r) for each unit cell in a SPUDT structure to obtain the desired P-matrix and, therefore, a desired frequency response. The measure of closeness between desired P-matrix and obtained P-matrix (computed P-matrix by Tabu Search) is measured by distance. The results obtained with Tabu Search are reasonably improved as compared to existing techniques such as Iterative Local Search Technique.