tors have also been presented. A convenient chart for filter design has been constructed using this model and the equivalent relations. A simple systematic design procedure has been established for filter designers. Finally, an example of a three-order compact filter has been designed. The new filter is much smaller than the conventional filter. This technique can also be applied to other planar structures, such as coplanar waveguide. This compact filter is very useful for MMIC design.
Abstract-A new class of microstrip filter structures are designed, optimized, simulated and measured for ultra-narrowband performance essential to the wireless industry applications. More accurate model of the coupling coefficient is outlined and tested for narrowband filter design. Two sample filters are fabricated and measured to verify the simulations and prove the concept. The idea behind the new designs is based on minimizing the parasitic couplings within the resonators and the inter-resonator coupling of adjacent resonators. A reduction of the overall coupling coefficient is achieved even with less resonator separation which is a major issue for compactness of such filters. The best new designs showed a simulated fractional bandwidth (F BW ) of 0.05% and 0.02% with separations of S = 0.63 mm and S = 0.45 mm, respectively. The measured filters tend to have even narrower F BW than the simulated, though its insertion loss deteriorates, possibly due to mismatch at the interface with external circuitry and poor shielding effect of the test platform. The investigated 2-pole filters are accommodated on a compact area of a nearly 0.6 cm 2 . An improvement of tens of times of order in narrowband performance is achieved compared to reported similar configuration filters and materials. A sharp selectivity and quasi-elliptic response are also demonstrated with good agreement in both simulations and measurements. In all filters, however, the study shows that the narrower the F BW , the larger the insertion loss (IL) and the worse the return loss (RL). This is confirmed by measurements.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.