An Efficient Analysis Method for LTCC Ridge Waveguide Bandpass Filters via Database Searching.

Low temperature cofired ceramic (LTCC) ridge waveguide filters are attractive candidates in implementing microwave filters in modern communication systems because of their competitive features, such as low loss, low cost, compact size, lightweight, high-power handling capability, wide spurious-free out-of-band response, and easy integration. However, optimization of LTCC ridge waveguide filters suffers from low speed or imperfect accuracy when the traditional full-wave simulation or the mode-matching method is utilized for analysis. This article makes research on an alternative analysis method based on the network theory for LTCC inline ridge waveguide filters. The method has high speed since it evaluates filter properties by simply making the cumulative product of transfer matrices of all constitutive laminated ridge waveguide (LRWG) sections and laminated coupling (LCP) sections according to their arranging sequence, and transfer matrices of all these sections are calculated from their characteristic parameters (transmission line or equivalent circuit parameters for LRWG or LCP sections) analytically. Another key feature of the method is that all characteristic parameters of LRWG and LCP sections are obtained by searching precreated database. In order to guarantee the analysis accuracy, two special strategies have been proposed and utilized. First, accurate extraction methods of characteristic parameters have been introduced and applied to the creation of a database for LRWG and LCP sections. Second, a piecewise interpolation scheme has been designed and adopted in the database searching process for both kinds of sections. The efficiency of the promoted analysis method has been verified by two analysis examples of LTCC LRWG filters. One is a three-order narrow-band filter and the other is a nine-order wide-band filter. The curves of scattering parameters calculated using the promoted method agree well with the simulated curves and the time consumed by the method is negligible when compared with the simulation time. The effectiveness of the proposed analysis method in filter design is also briefly introduced in this article. [ABSTRACT FROM AUTHOR]