Modeling Techniques for the efficient design of microwave bandpass filters

Abstract

Microwave bandpass filters are indispensable building blocks in modern telecommunication systems. They are designed to pass signals within certain frequency bands, while attenuating the spectral content that lies outside of these frequency bands. Over the last years the frequency spectrum has become more crowded, and this automatically led to more stringent filter specifications.
A filter is specified by a frequency template, which defines the position and the attenuation over the frequency bands that should be attenuated and passed. Several techniques have been developed in the state of the art to design a wide variety of microwave filters in various technologies including: waveguide, dielectric resonator and planar technologies such as microstrip filters.
Most designs must be optimized after an initial design process to meet the specifications. The main draw-back of current optimization methods is their need for many time-consuming electromagnetic (EM) simulations.
This work aims to reduce the filter design time using models to reduce the simulation count and assist the designer in time-efficient physical design of the microwave filters. To reach practical applicability, these models are used in different commonly used computer-aided design (CAD) frameworks.
Two different modeling strategies are developed, each with its own application goal. The first approach is based on coupling matrix theory and introduce the models in the design cycle of the filter itself. It provides physical insight in parasitic behavior, feeding the intuition of the designer. The method was also proved to improve and extend the design curves used to physically dimension the filter structure. The accuracy of the initial design is hereby again improved, and this reduces the required level of tuning. Moreover metamodels are also used to optimize filters for different design scenarios.
These methods have been applied to several state-of-the-art microstrip bandpass filter that were designed, realized and measured to show the applicability of the method in a state-of-the-art CAD environment.

Date of Award	19 Oct 2016
Original language	English
Awarding Institution	Electricity Université Côte d'Azur