Dissertations / Theses on the topic 'Filters’ design'

Thesis (M. Phil.)--Hong Kong University of Science and Technology, 2004.
Includes bibliographical references (leaves 108-111). Also available in electronic version. Access restricted to campus users.

Harmonics are created by non-linear devices connected to the power system. Power system harmonics are multiples of the fundamental power system frequency and these harmonic frequencies can create distorted voltages and currents. Distortion of voltages and currents can affect the power system adversely causing power quality problems. Therefore, estimation of harmonics is of high importance for efficiency of the power system network. The problem of harmonic loss evaluation is of growing importance for renewable power system industry by impacting the operating costs and the useful life of the system components. Non-linear devices such as power electronics converters can inject harmonics alternating currents (AC) in the electrical power system. The number of sensitive loads that require ideal sinusoidal supply voltage for their proper operation has been increasing. To maintain the quality limits proposed by standards to protect the sensitive loads, it is necessary to include some form of filtering device to the power system. Harmonics also increases overall reactive power demanded by equivalent load. Filters have been devised to achieve an optimal control strategy for harmonic alleviation problems. To achieve an acceptable distortion, increase the power quality and to reduce the harmonics hence several three phase filter banks are used and connected in parallel. In this thesis, high order harmonics cases have been suppressed by employing variants of Butterworth, Chebyshev and Cauer filters. MATLAB/SIMULINK wind farm model was used to generate and analyze the different harmonics magnitude and frequency. High voltage direct current (HVDC) lines for an electrical grid that is more than50km far away wind farm generation plant was investigated for harmonics. These HVDC lines are also used in offshore wind farm plant. Investigated three-phase harmonics filters are shunt elements that are used in power systems for decreasing voltage distortion and for correcting the power factor. Renewable energy sources are not the stable source of energy generation like wind, solar and tidal e.t.c. Though they are secondary sources of generation and hard to connect with electrical grid. In near future the technique is to use the wave digital filter (WDF) or circulator-tree wave digital filter (CTWDF) for the renewable energy application can be employed to mitigate the harmonics. These WDF and CTWDF can b eused in HVDC lines and smart grid applications. A preliminary analysis is conducted for such a study.

[ES] El creciente número de dispositivos intercambiando datos ha empujado a las empresas del sector espacial a utilizar bandas de frecuencia cada vez más altas, como Ku, K y Ka, ya que permiten emplear canales de frecuencia más anchos. A medida que disminuye la longitud de onda, el tamaño de los filtros se reduce y, por tanto, son más sensibles a las desviaciones de fabricación. Para compensar estos errores, es necesario emplear elementos de sintonía en la etapa de diseño. En este contexto presentamos una estrategia de diseño que permite incluir todos los factores no ideales, como elementos de sintonía o esquinas redondeadas, en las simulaciones finales de filtros y multiplexores. Una vez se han fabricado los filtros es necesario ajustar manualmente los elementos de sintonía hasta recuperar la respuesta objetivo. Sin embargo, para realizar esta tarea con éxito es necesario tener mucha experiencia previa y, aún así, conlleva un tiempo considerable. Por tanto, también proponemos un procedimiento de sintonización eficiente y sistemático que permite a cualquier persona, independientemente de su experiencia previa en sintonización, realizar esta tarea con éxito. Además del aumento de las tasas de transmisión, otros desafíos del sector espacial son reducir el tamaño y peso de sus componentes, así como dotarlos de capacidad dereconfiguración. Emplear dispositivos multifunción como filtros multibanda o dispositivos reconfigurables es una posible solución. En este contexto, proponemos una nueva familia de filtros multibanda en guía de ondas que puede adaptarse a las futuras necesidades del sector espacial. Con el mismo objetivo, también proponemos una familia de dispositivos reconfigurables de varios estados discretos que pueden modificar su comportamiento de forma remota.
[CA] El creixent nombre de dispositius intercanviant dades ha empés a les empreses del sector espacial a utilitzar bandes de freqüència cada vegada més altes, com Ku, K i Ka, ja que permeten fer servir canals de freqüència més amples. A mesura que la longitud d'ona disminueix, la mida dels filtres es redueix i, per tant, són més sensibles a les desviacions de fabricació. Per compensar aquests errors, és necessari fer servir elements de sintonia en l'etapa de disseny. En aquest context presentem una estratègia de disseny que permet incloure tots els factors no ideals, com a elements de sintonia o cantonades arrodonides, en les simulacions finals de filtres i multiplexors. Una vegada s'han fabricat els filtres és necessari ajustar manualment els elements de sintonia fins a recuperar la resposta objectiu. Però, per realitzar aquesta tasca amb èxit és necessari tenir molta experiència prèvia i, així i tot, comporta un temps considerable. Per tant, també proposem un procediment de sintonització eficient i sistemàtic que permet a qualsevol persona, independentment de la seua experiència prèvia en sintonització, realitzar aquesta tasca amb èxit. A més de l'augment de les taxes de transmissió, altres desafiaments de el sector espacial són reduir la mida i pes dels seus components, així com dotar-los de capacitat de reconfiguració. Emprar dispositius multifunció com filtres multibanda o dispositius reconfigurables és una possible solució. En aquest context, proposem una nova família de filtres multibanda en guia d'ones que pot adaptar-se a les futures necessitats del sector espacial. Amb el mateix objectiu, també proposem una família de dispositius reconfigurables de diversos estats discrets que poden modificar el seu comportament de forma remota.
[EN] The need for ever increasing data rate of modern communication systems has motivated companies in the space sector to exploit higher frequency bands, such as Ku, K and Ka, in order to offer wider bandwidths to their customers. However, as the frequency increases, the wavelength decreases, and all waveguide hardware becomes smaller and more sensitive to deviations from the ideal dimensions that normally occur when manufacturing the devices. In order to compensate for these deviations (or errors), tuning elements must then be added to the hardware and included in the design process. In this context, therefore, we focus on the investigation of novel design strategies for filters and multiplexers with the objective of including all necessary non-ideal factors in the design process. It is important to note in this context that, once the filters are manufactured, the tuning elements are usually adjusted manually until the desired target performance has been achieved. However, successfully performing this task requires a considerable amount of time and very significant previous experience in tuning microwave filters. Consequently, an additional goal of our research work is to propose efficient and systematic tuning procedures so that anyone, regardless of their previous tuning experience, can successfully perform this difficult task. In addition to the increasing data rates, another current challenge of advanced communication systems is the ability to be reconfigured remotely to adjust to changes in costumer demands. The use of multi-function or reconfigurable devices is then an attractive possible solution. In this context, therefore, we also investigate new families of multi-band waveguide filters that can be used to accommodate several pass bands in the same filtering device. Furthermore, we also propose a new family of reconfigurable devices with several discrete states that can be easily controlled remotely.
Melgarejo Lermas, JC. (2021). Advanced Techniques for the Design and Optimization of Multi-Band and Reconfigurable Microwave Waveguide Filters [Tesis doctoral]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/172728
TESIS

El desplegament mundial de la tecnologia 5G i l’ús estès de xarxes sense fils han impulsat un creixement accelerat del nombre de bandes freqüencials requerides pels telèfons intel·ligents i altres dispositius mòbils. De manera simultània, per incrementar la velocitat de connexió, també s’incrementen la complexitat de l’esquema de modulació, el nombre d’antenes i el nombre de portadores agregades. Aquestes especificacions requereixen avenços en la tecnologia de filtres d’RF per assegurar la selecció de la portadora adequada i de tota la informació transmesa. Alhora, dispositius més petits son necessaris en un espai cada cop més reduït i han de suportar nivells de potència més elevats en un escenari d’alta integració. Davant d’aquest desafiament, els ressonadors basats en tecnologia d’ona acústica han demostrat ser la solució que compleix amb els requeriments del mercat. El disseny d’un filtre d’ona acústica ha estat enfocat fins ara des de les tècniques d’optimització que requereixen un procés llarg i costós en termes de computació, una aproximació poc eficient des del punt de vista de la indústria. Però, eines de simulació basades en tècniques avançades de síntesi que consideren les restriccions tecnològiques han estat desenvolupades pel nostre grup en els darrers anys. Aquestes eines proporcionen solucions ràpides i precises com a primer pas per la posterior optimització de l’estructura. El propòsit principal d’aquesta tesi és aconseguir les demandes actuals d’integració de múltiples filtres en un sol mòdul (un multiplexor) i també incloure amplificadors de potència per reduir la mida de les capçaleres d’RF dels dispositius mòbils. En el cas dels multiplexors, el concepte de les xarxes de mínima susceptància ha estat empleat introduint el control de la fase del coeficient de reflexió per classificar aquestes xarxes. Dos escenaris han estat analitzats en quan al pla de freqüències: fixe i flexible. Aquest darrer és un cas complex que es troba impulsat en el paradigma de l’agregació de portadores i els requeriments del mercat. La metodologia proposada garanteix el compliment de les restriccions tecnològiques i dels requeriments d’atenuació de cada canal. En la part de la integració de filtres d’ona acústica en mòduls complets, aquesta tesi adreça la síntesi de filtres considerant impedàncies complexes a la càrrega i a la font. Aquesta proposta permet codissenyar filtres amb xarxes actives o passives permetent eliminar la tradicional etapa intermèdia d’adaptació d’impedàncies. El process de codisseny és il·lustrat amb amplificadors de potència i filtres analitzant les característiques de cara part per assegurar una alta eficiència i una alta potència de sortida en la banda. Alhora, una resposta de filtrat de Chebyshev s’obté en el dispositiu resultant. De la mateixa manera, el control de la fase amb diversos objectius s’ha discutit en aquesta tesi. L’anàlisi teòric queda suportat per exemples simulats i prototips fabricats que demostren la validesa de la hipòtesi. Els resultats aconseguits es troben resumits al final de cada capítol.
La implementación de la tecnología 5G a nivel mundial y el extendido uso de conexiones inalámbricas han impulsado el incremento acelerado del número de bandas de radio frecuencia (RF) que deben ser soportadas por los teléfonos inteligentes y los dispositivos móviles. Al mismo tiempo aumenta la complejidad de los esquemas de modulación, el número de antenas y la cantidad de portadoras con el objetivo de aumentar la velocidad de conexión. Estas demandas requieren avances en la tecnología de filtros de RF que garanticen la selección correcta de cada portadora y la adquisicion íntegra de la información deseada. Sumamos además la necesidad de tener dispositivos cada vez más pequeños en un espacio cada vez más reducido y capaces de manejar mayores potencias en un escenario de alta integración. En este contexto tan desafiante, los filtros basados en resonadores de tecnología microacústica han demostrado ser la solución para cubrir las necesidades del mercado. El diseño de filtros en esta tecnología ha estado dominado por técnicas de optimización que requieren un gran tiempo de procesado y un esfuerzo computacional no eficientes desde el punto de vista industrial. Sin embargo, herramientas de simulación basadas en métodos de síntesis avanzados que consideran los requerimientos propios de la tecnología han sido desarrollados en los últimos años en nuestro grupo de investigación con el objetivo de proporcionar una solución precisa y rápida que sirva de semilla para una posterior optimización. Esta tesis persigue como objetivo dar respuesta a la tendencia actual de integración de múltiples filtros en módulos de RF que incluyen amplificadores de potencia (PA) para disminuir la ocupación dentro de los smartphones. Durante el proceso de diseño de multiplexores he aplicado conceptos con fundamentos teóricos sólidos como las Redes de Mínima Susceptancia y he introducido el control de la fase en el proceso de síntesis como elemento clave en la clasificación de dichas redes de filtrado. Dos posibles escenarios fueron analizados según el plan de frecuencia: fijo y flexible. Este último escenario es especialmente complejo y constituye una demanda industrial reciente debido a la aparición de la tecnología de Carrier Aggregation. La metodología presentada garantiza la viabilidad tecnológica y el cumplimiento de las especificaciones para cada banda de frecuencia. Como parte de la integración en módulos, esta tesis también incluye la síntesis de filtros acústicos con impedancia compleja a la entrada y/o salida. Esta propuesta permite el codiseño de filtros con elementos activos o pasivos para eliminar redes intermedias de adaptacion de impedancia. Se ilustra el proceso de codiseño de amplificadores de potencia y filtros en tecnología microacústica desde un punto de vista integrador para garantizar una alta eficiencia en la conversión de energía y potencias de salida estables en toda la banda de frecuencia. Simultáneamente, una respuesta de filtrado tipo Chebyshev es obtenida en el dispositivo final. El control de la fase con diferentes objetivos es también abordado en este documento. Todos los análisis teóricos han sido acompañados de ejemplos simulados y prototipos fabricados que demuestran la ideoneidad de los planteamientos. Los resultados alcanzados han sido resumidos en cada capítulo.
The worldwide implementation of 5G technology and the extended use of wireless networks have boosted the fast-paced increase in the radio frequency bands number supported by smartphones and other mobile devices. Simultaneously, to increment the connection speed, the modulation scheme complexity, antenna number, and carrier aggregated are increasing. These specifications require advances in the RF filter technology to ensure the proper selection of each carrier and the full information acquisition. Additionally, smaller devices are needed for a shrinking space and also be able to handle higher powers in a highly integrated scenario. In this challenging environment, resonators based on acoustic wave (AW) technology have demonstrated to be the solution for the market requirements. The AW filter design has been driven by optimization techniques that require a time-consuming process and computational efforts, being inefficient from the industry point of view. Nevertheless, simulation tools based on advanced synthesis techniques considering the technological accommodation have been developed in our research group in the last years. These tools provide precise and fast solutions as the first tier for a later optimization procedure. The principal purpose of this thesis is to meet the current demands of integration of multiple filters in one RF module, namely multiplexers, and also including power amplifiers (PA) to reduce the device size inside the smartphones. For the multiplexers design, the Minimum Susceptance Networks (MSN) concept has been exploited, introducing the reflection coefficient phase control as a key element for this kind of network classification. Two probable scenarios were analyzed regarding the frequency plan: fixed and flexible. The latter is a complex case boosted by the advent of the Carrier Aggregation technology and the market requirements. The proposed methodology guarantees technological feasibility and mask specifications fulfillment for each channel. As part of the AW filters module integration, this thesis addresses the synthesis considering complex source/load impedances. The proposal allows the co-design of filters with active or passive networks removing the common inter-stage matching network. It is illustrated the co-design process for power amplifiers and filters analyzing each part's features to ensure high efficiency and output power in the passband. Simultaneously, a Chebyshev filter response is obtained in the resulting device. Moreover, the phase control with different purposes is also discussed in this document. The theoretical analysis has been supported by simulated examples and manufacturing prototypes that show the approach's pertinence. The achieving results have been summarized at each chapter's end.
Universitat Autònoma de Barcelona. Programa de Doctorat en Enginyeria Electrònica i de Telecomunicació

This thesis investigates multi-passband and tunable microwave filters, it includes a new generalised design technique for multi-passband filters and a new coupling tuning structure for tunable waveguide filters. The synthesis technique is an analytical approach and offers very fast solutions to the design once the desired filter specifications are given. The technique calculates the coupling matrix and external Q-factors for a wide range of filter specifications. The centre frequency and bandwidth of each passband, and the number of passbands can all be arbitrarily chosen. To verity the calculations, multi-passband filters are realised by using inverter coupled resonator sections. Two X-band waveguide multi-passband filter examples are given to validate the theory. Besides the innovation in synthesis technique, new designs of fully tunable waveguide filters are also presented. These filters are based on the new coupling tuning structure and a separate frequency tuning structure offering tuning in both centre frequency and bandwidth of the filter. One tunable bandpass filter and one tunable notch filter is implemented in X-band waveguide circuit.

The need for fast multidimensional signal processing arises in many areas. One of the more demanding applications is real time visualization of medical data acquired with e.g. magnetic resonance imaging where large amounts of data can be generated. This data has to be reduced to relevant clinical information, either by image reconstruction and enhancement or automatic feature extraction. Design of fast-acting multidimensional filters has been subject to research during the last three decades. Usually methods for fast filtering are based on applying a sequence of filters of lower dimensionality acquired by e.g. weighted low-rank approximation. Filter networks is a method to design fast multidimensional filters by decomposing multiple filters into simpler filter components in which coefficients are allowed to be sparsely scattered. Up until now, coefficient placement has been done by hand, a procedure which is time-consuming and difficult. The aim of this thesis is to investigate whether genetic algorithms can be used to place coefficients in filter networks. A method is developed and tested on 2-D filters and the resulting filters have lower distortion values while still maintaining the same or lower number of coefficients than filters designed with previously known methods.

El ràpid creixement dels sistemes de telecomunicacions sense fils ha portat a una creixent demanda de dispositius mòbils amb requisits cada vegada més estrictes per als filtres de microones incloent un millor rendiment, miniaturització i la reducció de costos. Una nova tecnologia de filtres de RF s’ha consolidat en l'última dècada: la tecnologia BAW. Els filtres BAW són perfectament adequats com a filtres miniaturitzats i d'alt rendiment per a aplicacions d'alta freqüència. En comparació amb la tecnologia SAW, els filtres BAW proporcionen menors pèrdues d'inserció, una millor selectivitat, compatibilitat CMOS, millor maneig de potència i una major freqüència de treball. Un CRF BAW és un dels més recents filtres BAW que permet oferir una àmplia gamma de noves i interessants propietats en comparació dels seus predecessors. Basat en el concepte de ressonadors acoblats acústicament, es poden oferir respostes Chebyshev clàssiques i avançades, millor ample de banda, més miniaturització i conversió de mode. La complexitat d'un CRF és considerable en termes de disseny i fabricació, a causa de la quantitat de làmines que el conformen. El disseny de filtres BAW es basa en dimensionar les capes del dispositiu. El disseny és complex i els procediments d'optimització poden no ser eficients a causa de la quantitat de variables que intervenen en el procés. En aquest treball es presenta una metodologia d'anàlisi i síntesi. Les tècniques poden fer front a les estructures de diverses capes d'una manera directa. Es presenta un conjunt de models circuitals completament elèctrics que simplifiquen i recullen el comportament dels CRF, dispositius que funcionen tant en els dominis elèctric i mecànic. S'han establert un conjunt d'eines per analitzar el ressonador per mitjà de teoria de xarxes i és la base per a l'anàlisi d'estructures compostes de diversos ressonadors i múltiples capes de làmines primes. Es desenvolupa un model per al CRF utilitzant transformacions de xarxa i aproximacions respecte a les freqüències de funcionament dels dispositius BAW. Es presenta un anàlisi profund del BAW CRF amb els models proposats com una eina que permet guiar el procediment analític. L'anàlisi dels filtres facilita una millor comprensió del seu funcionament i una manera d'establir els límits de les respostes que poden oferir. Es presenten metodologies de disseny de filtres amb la finalitat d'obtenir les geometries de l'estructura i la topologia dels filtres BAW per complir amb les respostes prescrites. El disseny del filtre es dona a partir d'una comprensió clara dels seus mecanismes de filtratge de manera que facilita l'explotació de les seves característiques inherents. Les respostes avançades amb zeros de transmissió o els filtres de banda de rebuig procedents dels nous dispositius presentats es sintetitzen per mitjà de les metodologies proposades.
El rápido crecimiento de los sistemas de telecomunicaciones inalámbricas ha llevado a una creciente demanda de dispositivos móviles con requisitos cada vez más estrictos para los filtros de microondas incluyendo un mejor rendimiento, miniaturización y la reducción de costes. Una nueva tecnología de filtros de RF se ha consolidado en la última década: la tecnología BAW. Los filtros BAW son perfectamente adecuados como filtros miniaturizados y de alto rendimiento para aplicaciones de alta frecuencia. En comparación con la tecnología SAW, los filtros BAW proporcionan menores pérdidas de inserción, una mejor selectividad, compatibilidad CMOS, mejor manejo de potencia y una mayor frecuencia de trabajo. Un CRF BAW es uno de los más recientes filtros BAW que permite ofrecer una amplia gama de nuevas e interesantes propiedades en comparación con sus predecesores. Basado en el concepto de resonadores acoplados acústicamente, se pueden ofrecer respuestas Chebyshev clásicas y avanzadas, mejor ancho de banda, más miniaturización y conversión de modo. La complejidad de un CRF es considerable en términos de diseño y fabricación, debido a la cantidad de láminas que lo conforman. El diseño de filtros BAW se basa en dimensionar las capas del dispositivo. El diseño es complejo y los procedimientos de optimización pueden no ser eficientes debido a la cantidad de variables que intervienen en el proceso. En este trabajo se presenta una metodología de análisis y síntesis. Las técnicas pueden hacer frente a las estructuras de varias capas de una manera directa. Se presenta un conjunto de modelos circuitales completamente eléctricos que simplifican y recogen el comportamiento de los CRF, dispositivos que funcionan tanto en los dominios eléctrico y mecánico. Se han establecido un conjunto de herramientas para analizar el resonador mediante teoría de redes y es la base para el análisis de estructuras compuestas de varios resonadores y múltiples capas de láminas delgadas. Se desarrolla un modelo para el CRF utilizando transformaciones de red y aproximaciones respecto a las frecuencias de funcionamiento de los dispositivos BAW. Se presenta un análisis profundo del BAW CRF con los modelos propuestos como una herramienta que permite guiar el procedimiento analítico. El análisis de los filtros facilita una mejor comprensión de su funcionamiento y una manera de establecer los límites de las respuestas que pueden ofrecer. Se presentan metodologías de diseño de filtros con el fin de obtener las geometrías de la estructura y la topología de los filtros BAW para cumplir con las respuestas prescritas. El diseño del filtro se da a partir de una comprensión clara de sus mecanismos de filtrado de modo que facilita la explotación de sus características inherentes. Las respuestas avanzadas con ceros de transmisión o los filtros de banda de rechazo procedentes de los nuevos dispositivos presentados se sintetizan por medio de las metodologías propuestas.
The speedy growth of wireless telecommunication systems has led to an increasing demand for hand-held devices with more and more stringent requisites for microwave filters including better performance, miniaturization and reduced costs. A new RF filter technology has emerged during the last decade: BAW technology. BAW filters are perfectly suitable as miniaturized high performance filters for high frequency and power applications. Compared to SAW filter technology, BAW filter solutions can provide lower insertion loss, better selectivity, CMOS compatibility, higher power handling and higher operation frequency.A BAW CRF is one of the latest BAW filter topologies to offer a range of new and interesting properties compared to its predecessors. Based on the concept of acoustically coupled resonators, it can provide classical and advanced Chebyshev responses, better bandwidths, more miniaturization and mode conversion. The complexity of a CRF is considerable in terms of design and fabrication due to the quantity of films that it comprises.The design of BAW filters is based on sizing the layers of the physical device. It becomes complicated with this amount of layers, and optimization procedures may not be efficient due to the number of variables involved in the process. A methodology to analyze and synthesize BAW CRFs is presented in this work. The techniques can deal with multilayered structures in a straightforward way.A set of fully electric circuital models that simplify and gather the behavior of CRFs, devices that work both in the electrical and the mechanical domains, are presented. A set of tools to analyze the resonator by means of network theory is established and is the basis for analyzing structures composed of several resonators and multiple thin-film layers. A model for the most basic CRF is developed using network transformations and approximations regarding the frequencies of operation of the BAW devices.A profound analysis of BAW CRFs is presented using the proposed models as an enabling tool to guide the analytical procedure. The analysis of those filters facilitates a better understanding of their performance and a way to establish the limits on the responses that they can provide. Systematic filter design methodologies are presented in order to obtain the structure geometries and topologies of BAW filters to fulfill prescribed responses in such a way that fully time-consuming and sometimes not very controlled optimization-oriented procedures are avoided. The design of the filter from a clear understanding of its performances facilitates the exploitation of its inherent characteristics. Advanced responses with transmission zeros or stopband filters coming from new devices are synthesized by means of the proposed methodologies.

Optical filters encompass a vast array of devices and structures for a wide variety of applications. Generally speaking, an optical filter is some structure that applies a designed amplitude and phase transform to an incident signal. Different classes of filters have vastly divergent characteristics, and one of the challenges in the optical design process is identifying the ideal filter for a given application and optimizing it to obtain a specific response. In particular, it is highly advantageous to obtain a filter that can be seamlessly integrated into an overall device package without requiring exotic fabrication steps, extremely sensitive alignments, or complicated conversions between optical and electrical signals. This dissertation explores three classes of nano-scale optical filters in an effort to obtain different types of dispersive response functions. First, dispersive waveguides are designed using a sub-wavelength periodic structure to transmit a single TE propagating mode with very high second order dispersion. Next, an innovative approach for decoupling waveguide trajectories from Bragg gratings is outlined and used to obtain a uniform second-order dispersion response while minimizing fabrication limitations. Finally, high Q-factor microcavities are coupled into axisymmetric pillar structures that offer extremely high group delay over very narrow transmission bandwidths. While these three novel filters are quite diverse in their operation and target applications, they offer extremely compact structures given the magnitude of the dispersion or group delay they introduce to an incident signal. They are also designed and structured as to be formed on an optical wafer scale using standard integrated circuit fabrication techniques. A number of frequency-domain numerical simulation methods are developed to fully characterize and model each of the different filters. The complete filter response, which includes the dispersion and delay characteristics and optical coupling, is used to evaluate each filter design concept. However, due to the complex nature of the structure geometries and electromagnetic interactions, an iterative optimization approach is required to improve the structure designs and obtain a suitable response. To this end, a Particle Swarm Optimization algorithm is developed and applied to the simulated filter responses to generate optimal filter designs.
Ph.D.
Optics and Photonics
Optics and Photonics
Optics PhD

Product families are frequently used to provide consumers with a variety of appealing products and to help maintain reasonably low production costs for manufacturers. Three common objectives in the design of product families are used to balance the interests of both consumers and manufacturers. These objectives are to maximize (i) product performance, (ii) product distinctiveness as perceived by the consumer, and (iii) product commonality as seen by the manufacturer. In this thesis, three methods are introduced that use multiobjective optimization and Smart Pareto filtering to satisfy the three objectives of product family design. The methods are progressive in nature and begin with the selection of product family members using Smart filtering and develop through the establishment of scale- based product platforms to the design of combined scale- based and module-based product platforms. Each of the methods is demonstrated using a well-know universal electric motor example problem. The results of each method are then compared to a benchmark electric motor product family that was previously defined in the literature. Additionally, a pressure vessel example problem is used to further demonstrate the first of the three methods.

Most signal processing / communications applications heavily rely on filters. For adaptive spectrum filtering and for applications that switch between sets of different filter implementations, it would be beneficial to utilize just one, tuneable band-pass filter. In recent years, the study of metamaterials emerged as an area of scientific research due to the unique attributes of metamaterials. Metamaterials typically are artificial structures with properties not found in nature, for instance negative refraction indexes. Their feature sizes span a fraction of the wavelength corresponding to their frequency of operation. A sub group of metamaterials, the electromagnetic bandgap (EBG) structures, exhibit stopbands for electromagnetic waves irrespective of polarization or angle of incidence. EBG structures prominently achieved surface wave suppression to minimise cross talk between neighbouring devices and improving antenna efficiency by acting as a perfect magnetic conductor within a certain frequency range. The thesis investigates the suitability of EBG structures for filter implementations. The goal is to provide a tuneable band-pass filter for adaptive spectrum filtering and communication applications. The bandgap of an infinite array of EBG cells is numerically determined. Based on those results, an EBG band-pass filter implementation on a printed circuit board (PCB) is designed, fabricated and characterized. Different tuning methods were incorporated into the PCB design to create a tuneable EBG band-pass filter. An EBG filter was built on a fused silica wafer, in order to shift the passband to higher frequencies.

The design of microwave filters starts from the derivation of a defined lowpass prototype network. A general lossy synthesis method is given which can 1) derive the reflection function from the transfer function when the unitary condition is not satisfied; 2) find the expressions for the complex admittance parameters and 3) synthesize the lossy coupling matrix (CM) with prescribed loss distributions. Two special cases are discussed for solving the refection function from a prescribed transfer function. An alternative approach to cope with loss is studied. In a transversal array, some resonators can be replaced by their low-Q alternatives to reduce the manufacture cost as well as the cavity size. The exact values for the dissipations of resonators or couplings can be determined analytically or by methods of gradient based optimizations. A method of CM synthesis with non-ideal load is given which can be used in designing diplexers or multiplexers. Filter networks matching to complex load impedances can be found by renormalizing reference impedances. An iteration method is introduced which can deal with frequency variant load and can deliver the required reflection zeros. A method for the synthesis of directional filters is presented which can be used for designing combiners. While each section of directional filters provides a 1st order response, more complex filter characteristics can be realized by cascading those single sections. By proper transformations, directional filter networks can be realized using normal resonators and couplings. An example utilizing coaxial resonator is given. A method for the analysis of 2-D lumped element networks is presented. The method is based on the general telegrapher’s equations of multi-wire transmission lines. A 2-D lumped element network is equivalent to a combination of sub-networks which support single mode propagations. The method can be applied to the analysis of metamaterials and can be used for the design of waffle-iron filters.

The design of envelope-constrained (EC) filters is considered for the time-domain synthesis of filters for signal processing problems. The objective is to achieve minimal noise enhancement where the shape of the filter output to a specified input signal is constrained to lie within prescribed upper and lower bounds. Traditionally, problems of this type were treated by using the least-square (LS) approach. However, in many practical signal processing problems, this "soft" least-square approach is unsatisfactory because large narrow excursions from the desired shape occur so that the norm of the filter can be large and the choice of an appropriate weighting function is not obvious. Moreover, the solution can be sensitive to the detailed structure of the desired pulse, and it is usually not obvious how the shape of the desired pulse should be altered in order to improve on the solution. The "hard" EC filter formulation is more relevant than the "soft" LS approach in a variety of signal processing fields such as robust antenna and filter design, communication channel equalization, and pulse compression in radar and sonar. The distinctive feature is the set of inequality constraints on the output waveform: rather than attempting to match a specific desired pulse, we deal with a whole set of allowable outputs and seek an optimal point of that set.The EC optimal filter design problems involve a convex quadratic cost function and a number of linear inequality constraints. These EC filtering problems are classified into: discrete-time EC filtering problem, continuous-time EC filtering problem, and adaptive discrete-time EC filtering problem.The discrete-time EC filtering problem is handled using the discrete Lagrangian duality theory in combination with the space transformation function. The optimal solution of the dual problem can be computed by finding the limiting point of ++
an ordinary differential equation given in terms of the gradient flow. Two iterative algorithms utilizing the simple structure of the gradient flow are developed via discretizing the differential equations. Their convergence properties are derived for a deterministic environment. From the primal-dual relationship, the corresponding sequence of approximate solutions to the original discrete-time EC filtering problem is obtained.The continuous-time EC filtering problem (semi-infinite convex programming problem) is handled using the continuous Lagrangian duality theory and Caratheodory's dimensionality theory. Several important properties are derived and discussed in relation to practical engineering requirements. These include the observation that the continuous-time optimal filter via orthonormal filters has the structure of a matched filter in cascade with another filter. Furthermore, the semi-infinite convex programming problem is converted into an equivalent finite dual optimization problem, which can be solved by the optimization methods developed. Another issue, which relates to the continuous-time optimal filter design problem, is the design of robust optimal EC filters. The robustness issue arises because the solution of the EC filtering problem lies on the boundary of the feasible region. Thus, any disturbance in the prescribed input signal or errors in the implementation of the optimal filter are likely to result in the output constraints being violated. A detailed formulation and a corresponding design method for improving the robustness of optimal EC filters are given.Finally, an adaptive algorithm suitable for a stochastic environment is presented. The convergence properties of the algorithm in a stochastic environment are established.

Yes
This study presents an original asymmetric stepped-impedance resonator filter combined with meander coupled-line structures and enabling the realisation of finite transmission zeros (TZs) and the implementation of multi-band bandpass filters. The meander coupled sections (MCSs) tune the TZs and resonant frequencies: with higher-order spurious frequencies cancelled by the TZs, a single wideband with wide stopband from 1.18 to 1.84 GHz is possible. Furthermore, by positioning the finite TZs between the high-order spurious frequencies and adjusting the coupling strength between resonators, a quint-wideband filter can be realised, with centre frequencies of 1.19, 4.29, 5.43, 6.97, 9.9 GHz and fractional bandwidths of 31.9, 15.4, 15.8, 4.3, 39.2%, respectively. More importantly, two filters with single/quad-wideband performance can be realised by tuning the parameters of the MCS, and therefore they can be designed separately by using only one original structure. The triple-wideband filter is realised with the help of the asymmetric parallel uncoupled microstrip section. These filter structures enjoy the advantage of single/multi-band versatility, structure reusability and simplicity. The good in-band and out-of-band performance, low loss and simple structure of the proposed single/tri/quint-wideband filters make them very promising for applications in future multi-standard wireless communication.
European Union's Horizon 2020 research and innovation programme under Grant agreement H2020-MSCA-ITN-2016 SECRET-722424.

博士
國立臺灣大學
電機工程研究所
82
The work of this dissertation is to devise novel and efficient techniques for optimally designing one-dimensional (1-D) digital filters, two-dimensional (2-D) digital filters, 1-D multirate filter banks, and 2-D multirate filter banks in the appropriate optimal sense. As for the 1-D digital filters, the proposed techniques include the design algorithms for designing quasi-equiripple FIR and IIR digital filters, discrete coefficient FIR digital filters with arbitrary amplitude and phase response, digital all-pass filters, high order digital differentiators using $L_{1}$ error criteria, sharp FIR digital filters with prescribed group delay phase response, and cascade form FIR digital filters with powers-of-two coefficients in the complex domain, and Chebyshev design of IIR digital filters with arbitrary magnitude and phase responses. On the other hand, this dissertation also presents two approaches for equiripple design of 2-D linear-phase FIR digital filters. One approach is based on a novel minimax design method and the other approach is the McClellan transform based design techniques. With regard to the 1-D filter bank systems, a deep study of designing quadrature mirror filters (QMF) with linear phase in the frequency domain using different optimal criteria is given. The powers-of-two coefficient design of QMF bank is considered, too. Other than the uniform-division QMF banks, the design of two-channel nonuniform-division maximally decimated filter banks is thoroughly studied. Considering the design of 2-D multirate filter bank systems, a novel minimax design of 2-D nonseparable QMF banks with non-diagonal decimation/ expansion matrix is presented. It is shown that the quincunx QMF bank and the parallelogram QMF bank can be easily designed using the method.

碩士
國立臺灣大學
電信工程學研究所
92
In this thesis, we consider three kinds of digital filters with filter coefficients taking on -1, 0, +1 only. One is variable fractional delay FIR filter and the others are IIR allpass filter and filter banks. We design these two kinds of digital filters with continuous coefficients using WLS algorithm and Karmarkar algorithm, in minmax criteria methods are proposed.. When we implement a filter of conventional structure, we always need multibit multipliers. But the circuit complexity and high cost of multibit multipliers always confuse us. Thus, we adopt a new filter structure [25] whose main part consists of a transversal filter with tap coefficients restricted to -1, 0, +1 only and cascaded with an appropriate recursive network with some specific resetting function. Therefore, it is not necessary for transversal filter to use multipliers; the adopted configuration is suitable for hardware implementation. We apply the new structure to the design of the digital filter we discuss above. In this thesis, we design variable delay FIR filter, IIR allpass filter, and IIR filter banks. The design examples are demonstrated to illustrate the effectiveness of the new structure.

博士
國立臺灣科技大學
電子工程技術研究所
86
This dissertation presents several novel and efficient techniques for optimally designing one-dimensional (1-D) perfect-reconstruction (PR) filter banks, two-dimensional (2-D) FIR digital filters, and 2-D perfect- reconstruction and near-perfect-reconstruction parallelogram filter banks in the minimax senses. The proposed approaches are developed based on the affine and dual affine scaling variants of Karmarkar''s algorithm. As for the 1-D perfect-reconstruction digital filter banks, two novel techniques are proposed for designing PR filter banks with FIR analysis and synthesis filters having linear phase responses as well as low delay characteristics. The designed analysis and synthesis filters for both cases are optimal in the minimax sense subject to the perfect-reconstruction constraints. With regard to the design of 2-D digital filters, we propose design techniques for continuous and powers-of-two coefficients 2-D digital filters based on the minimax sense. The optimal continuous coefficient filters are first designed by an affine scaling variant of Karmarkar''s algorithm. Then a suboptimal powers-of-two coefficient filter is obtained by an efficient method from the optimal continuous filter coefficients. The design of 2-D parallelogram filter banks is also studied thoroughly. The linear-phase FIR analysis and synthesis optimal in minimax sense are considered. Two novel techniques for designing perfect-reconstruction and near-perfect-reconstruction 2-D parallelogram filter banks are presented. From the simulation examples demonstrated in each chapter of this dissertation, the effectiveness of the proposed design techniques for each considered design problem can be confirmed.

碩士
國立中正大學
電機工程研究所
99
There are two parts in this thesis. The first part is design of a planar microwave dual-band bandpass filter. The structure of this dual-band bandpass filter is composed of an open-end coupled line and two parallel coupled lines with coupling feeding. Because the unequal even-mode and odd-mode phase velocity of the open-end coupled line, the interdigital capacitors at two ends of open-end coupled line are utilized to compensate the odd-mode phase velocity. Moreover, the added capacitive coupling between the input and output ports can move two transmission zeros at the passband skirts more closely. Then, two of single passband filters with different central frequencies are cascoded to achieve a dual-passband bandpass filter, and the J-inverter has been used to analyze this filter. In the second part, the planar microstrip bandpass filters with wide stopband are presented. The open-end coupled line, transmission line and short stub are the main structure of the wide stopband bandpass filter. By properly choosing electrical lengths of open-end coupled line, transmission line and short stub, the outband transmission zeros can be controlled to achieve a wide stopband. Moreover, the J-inverter can be used to analyze this filter. In addition, in order to increase the stopband bandwidth, two spur lines can be added into the input and output ports. All filters in this thesis are simulated with full-wave electromagnetic simulator, and these filters have been fabricated and measured. The matched results between electromagnetic simulation and measurement can demonstrate the availability of all proposed circuits.

Multidimensional (MD) multirate systems, which find applications in the coding and compression of image and video data, and in high definition television (HDTV) systems, have recently attracted much attention. Central to these systems is the idea of sampling lattices. The basic building blocks in an MD multirate system are the decimation matrix M, the expansion matrix L, and MD digital filters. When M and L are diagonal, most of the one-dimensional (1D) multirate results can be extended automatically, using separable approaches (i.e., separate operations in each dimension). Separable approaches are commonly used in practice due to their low complexity in implementation. However, nonseparable operations, with respect to nondiagonal decimation and expansion matrices, often provide more flexibility and better performance. Several applications, such as the conversion between progressive and interlaced video signals, actually require the use of nonseparable operations. For the nonseparable case, extensions of 1D results to the MD case are nontrivial. In this thesis, we will introduce some developments in these extensions. The three main results are: the design of nonseparable MD filters and filter banks derived from 1D filters, the commutativity of MD decimators and expanders and its applications to the efficient polyphase implementation of MD rational decimation systems, and the vector space framework for unifying MD filter bank and wavelet theory. In particular, properties of integer matrices like matrix fraction descriptions, coprimeness, the Bezout identity, etc., of which the polynomial versions are known in system theory, are used for the first time in the area of multirate signal processing.

碩士
國立中央大學
電機工程學系
104
This thesis used PIN diodes as the on/off switching component and utilized different pathways and boundary conditions to switch between two or more states on the same circuit. This study primarily uses band-stop filters (BSF) as the core of the passive circuitry along with ultra-wideband pass filters (UWBPF), narrow band-pass filters (NBPF), or multi-band stop circuitry to realize switchable circuits using specific design techniques. Since the BSF is the main component in the design of the switchable circuit and its structure is very similar to that of the UWBPF, an entire chapter is dedicated to its introduction and analysis. In the present study, a total of four circuits are detailed. The first circuit switches between a BSF and an UWBPF. The circuit can be further divided into designs with multiple stages (two-stage or three-stage) and according to the size of the bandwidth for a total of six circuit designs. The second circuit (a single/dual-band stop filter switchable design) implements a switching design, which leverages the fact that the electrical length of the BSF can be used to adjust the center frequency and the symmetry properties of the stop band. The third circuit (a switching circuit for BSF and NBPF) is created by first converting the BSF to a design with two subcircuits in parallel and then utilizing the open stub in the middle to implement the NBPF. The last circuit uses the fact that NBPFs have different feed-in methods to implement switching. FR4 PCBs with a thickness of 1.6 mm, dielectric constant of 4.4, and a loss tangent of 0.0245 were used for all implementations.

碩士
國立臺灣大學
電信工程學研究所
91
In this study, new low-pass, high-pass, and bandpass coplanar-waveguide (CPW) filters are implemented and carefully examined. First, we use various series and shunt stubs to accomplish a novel low-pass filter. The required stub structures are realized by properly etching on both signal stirp and ground planes of CPW. By periodically cascading this low-pass structure, we can improve the characteristics in the stopband. Second, a simple high-pass filter is implemented in the same manner as in low-pass one. The measured and simulated results of the low-pass and high-pass filters are compared and good agreement between them is observed. Finally, we cascade the filter structures mentioned above to realize a CPW bandpass filter. By properly designing the associated low-pass and high-pass structures, one may control the performance of the cascade bandpass filter. The design principle of achieving this bandpass circuit is described and also confirmed by the experimental and theoretical results, which are in good agreement. These proposed planar filters are easy to fabricate and compact, and they are suitable for communication systems.

Design of filters used in grid-connected inverter applications involves multiple constraints. The filter requirements are driven by tight filtering tolerances of standards such as IEEE 519-1992–IEEE Recommended Practices and Requirements for Harmonic Control in Electrical Power Systems and IEEE 1547.2-2008–IEEE Application Guide for IEEE Std 1547, IEEE Standard for Interconnecting Distributed Resources with Electric Power Systems. Higher order LCL filters are essential to achieve these regulatory standard requirements at compact size and weight. This objective of this thesis report is to evaluate design procedures for such higher order LCL filters. The initial configuration of the third order LCL filter is decided by the frequency response of the filter. The design equations are developed in per-unit basis so results can be generalized for different applications and power levels. The frequency response is decided by IEEE specifications for high frequency current ripple at the point of common coupling. The appropriate values of L and C are then designed and constructed. Power loss in individual filter components is modeled by analytical equations and an iterative process is used to arrive at the most efficient design. Different combinations of magnetic materials (ferrite, amorphous, powder) and winding types (round wire, foil) are designed and tested to determine the most efficient design. The harmonic spectrum, power loss and temperature rise in individual filter components is predicted analytically and verified by actual tests using a 3 phase 10 kW grid connected converter setup. Experimental results of filtering characteristics show a good match with analysis in the frequency range of interconnected inverter applications. The design process is stream-lined for the above specified core and winding types. The output harmonic current spectrum is sampled and it is established that the harmonics are within the IEEE recommended limits. The analytical equations predicting the power loss and temperature rise are verified by experimental results. Based on the findings, new LCL filter combinations are formulated by varying the net Lpu to achieve the highest efficiency while still meeting the recommended IEEE specifications. Thus a design procedure which can enable an engineer to design the most efficient and compact filter that can also meet the recommended guidelines of harmonic filtering for grid-connected converter applications is established.

Thesis (M.S.)--University of Tennessee, Knoxville, 2002.
Title from title page screen (viewed Sept 25, 2002). Thesis advisor: L. Montgomery Smith. Document formatted into pages (viii, 135 p. : ill.). Vita. Includes bibliographical references (p. 66-68).

碩士
國立東華大學
電機工程學系
102
In modern VLSI design, especially in system-on-chip, the number of transistors in a single chip keeps increasing thanks to the advance of chip manufacturing technology. However, as the feature size of modern chips shrinks, the circuits become more and more susceptible to noise, wire delay, and soft errors. One of these main problems is timing errors which are caused by process variation, device aging, etc. Such timing error problems can cause system failures. Hence, it is an important issue to solve the timing error problem while maintaining the performance of a chip. This thesis proposes various transformation designs for VLSI digital filters for tolerating multiple timing errors. We have developed a design methodology for VLSI digital filters, which can detect and tolerate multiple timing errors on-line. In order to achieve high performance of the digital filters, different transformations for various digital filter designs are applied. According to the design requirements, we choose the appropriate transformation for the filter in order to improve the performance, while it can still tolerate multiple timing errors. We have applied our techniques to two example digital signal filter designs, including a FIR filter and an IIR filter. Four examples for each circuit are studied and evaluated. We have implemented them using cell-based design flow on TSMC manufacturing technology. The implementation results show that our designs achieve high performance and tolerance of multiple timing errors for digital filters with reasonable cost.

碩士
國立臺灣科技大學
電子工程系
87
This thesis presents several novel and efficient techniques for designing three-dimensional (3-D) perfect reconstruction (PR) filter banks, FIR digital all-pass filters, and complex FIR digital filters in minimax sense. The proposed approaches are developed based on the affine and dual affine scaling variants of Karmarkar's algorithm. As for the 3-D perfect reconstruction digital filter banks, two novel techniques are proposed for designing PR filter banks with FIR analysis and synthesis filters having linear phase responses. The designed analysis and synthesis filters are in the minimax sense subject to the perfect reconstruction constraints. With regard to the design of FIR digital all-pass filters, we propose design techniques via minimizing the peak magnitude error and peak phase error simultaneously or minimizing the peak magnitude error and peak group delay error simultaneously. The filter coefficients are obtained by an affine scaling variant of Karmarkar's algorithm. For designing complex FIR digital filters, the original complex approximation is divided into two real ones first. Then the affine scaling variant of Karmarkar's algorithm is also applied to minimize the real part and imaginary part error in minimax sense to get the complex impulse response coefficients. From the simulation examples demonstrated in each chapter of this thesis, the effectiveness of the proposed design techniques for each considered problem can be confirmed.

碩士
淡江大學
電機工程學系
86
Power filters, in general, are the most economical and effective implements to solve the interference problems of harmonic in power sysems. Traditiionally, the engineers designed the filters by trial and error approaches. In these approaches, they did not consider the cost of the filters, the line loss, and ignorde the case that the tuned point is changed with the environment variations. So it makes a great reduction in the total performance. In this thesis, we propose a approach that is to use the optimum method to design the harmonic filters to limit the harmonic with the permissible maximal value (e. g. IEEE Std. 519). The proposed method minimizes the designed filter's total investment cost and power loss such that the harmonic distortion in within an acceptable range. 　　The passive filter, especially the single-tune filter, usually used to eliminate the harmonic in the power systems. It was widely accepted because of its simple configuration and the convenience of setting and maintenance. From the frequency point of view, the subject to design the filter is to prevent the harmonic currents(or voltages), except the base frequency current, flowing into the system. Traditionally, the engineers designed the filters with their experience. Although effective in eliminating the harmonic, these approaches did not consider the constraints on the environment and the operation. So the filters did not well work as expectation, sometimes they even caused series or parallel resonance to interfere other power equipments. 　　We address these problems from the viewpoints of the real operational condition and consider the cost of the filters and power loss with minimum. Simulation results obtained from an industrial distribution system demonstrate the effectiveness of the proposed method.

碩士
國立臺北科技大學
電腦與通訊研究所
98
The study of this thesis is to realize Composite Right/Left-Handed (CRLH) transmission line by using Complementary Split-Rings Resonators (CSRR). Applying this transmission feature to a passive circuit, an asymmetric fed structure of broadband bandpass filter with transmission zeros could be built with smaller size and better function. The designed small filter can be embedded in an antenna or other passive circuits, is expected to improve the overall performance of selectivity. Ansoft High Frequency Structure Simulation (HFSS) and IE3D are applied for simulations with Finite Element Method (FEM). Transforming the calculation of S parameters through MATLAB, it can be show that the refractive index (n), permittivity (ε), and permeability (μ) are all negative with the feature of Metamaterial. Deriving the equivalent circuit model of metamaterial bandpass filters from the equivalent material theory with the simulation software of Advanced Design System (ADS), Moment Method (MOM), and Momentum；We can accurately predict the S parameters with phase variations for left-handed materials further analysis and design.

碩士
國立高雄大學
電機工程學系碩士班
98
In this thesis, the major research is the design of variable digital filters. During the past decade, variable digital filters have received considerable attention because they were widely used in communication systems, signal processing and image processing. Different responses or delays can be immediately obtained by tuning the variable parameters without the need to design a new one. The designs of variable digital filters are often classified into two categories according to various adjustable situations, which are the variable fractional-delay (VFD) digital filters and the variable fractional-order (VFO) differintegrators. First, Weighted Least-Squares Approach (WLS) is proposed to design the FIR, IIR and Allpass variable fractional-delay filters. This method not only can achieve the desired performance but also minimize the errors in the design range. Then, the topic is focused on the variable fractional-order (VFO) differintegrators. Both FIR differintegrator and IIR differintegrator are designed. They can deal with derivatives or integrals or even compute derivatives and integrals in the same filter by simply adjusting proper parameters. Importantly, the Farrow structure has successfully been applied to implement the variable digital filter systems stated above.

碩士
國立高雄大學
電機工程學系碩士班
96
Variable digital filters have received considerable attention in the last decade due to their wide usages in communication systems and image processing. Their important advantage is that they contain a number of parameter in transfer function, which can be used to tune the frequency response immediately on-line without redesigning a new filter. The design of variable digital filters can be classified into two categories, the variable delay response filters and variable magnitude response filters. First, in terms of variable delay response filters, this thesis proposes a new coefficient relationship method to design variable fractional-delay (VFD) FIR filters, which can achieved higher designed performance in large order subfilters and less number of designed coefficient than the latest method. Then, the allpass IIR filters are also introduced to design VFD filters. Second, in the variable magnitude response filters, we propose two methods, binomial series expansion and Taylor series expansion, to design the variable fractional-order differentiators and integrators. Furthermore, in the design of 2D variable digital filters, the variable 2-D subfilters and variable 1-D prototype filters are designed with the same parameter to construct all kinds of variable 2-D FIR filters by using McClellan transformation. They can also be implemented by the proposed structures. Finally, the variable window is also generated which can be applied to design variable transition bandwidth FIR filters. In the thesis, several series expansions are used such that elements of relative matrices can be evaluated in closed-forms and annoying numerical integrals can be all avoided. Then, to find the optimal solution, the weighted least-squares (WLS) approach is used.

The main purpose of this dissertation was to address key issues in the design and analysis of RF/microwave filters for wireless applications. Since RF/microwave filters are one of the bulkiest parts of communication systems, their miniaturization is one of the most important technological challenges for the development of compact transceivers. In this work, novel miniaturization techniques were investigated for single-band, dual-band, ultra-wideband and tunable bandpass filters. In single-band filters, the use of cross-shaped fractals in half-mode substrate-integrated-waveguide bandpass filters resulted in a 37 percent size reduction. A compact bandpass filter that occupies an area of 0.315 mm2 is implemented in 90-nm CMOS technology for 20 GHz applications. For dual-band filters, using half-mode substrate-integrated-waveguides resulted in a filter that is six times smaller than its full-mode counterpart. For ultra-wideband filters, using slow-wave capacitively-loaded coplanar-waveguides resulted in a filter with improved stopband performance and frequency notch, while being 25 percent smaller in size. A major part of this work also dealt with the concept of 'hybrid' RF MEMS tunable filters where packaged, off-the-shelf RF MEMS switches were used to implement high-performance tunable filters using substrate-integrated-waveguide technology. These 'hybrid' filters are very easily fabricated compared to current state-of-the-art RF MEMS tunable filters because they do not require a clean-room facility. Both the full-mode and half-mode substrate-integrated waveguide tunable filters reported in this work have the best Q-factors (93 - 132 and 75 - 140, respectively) compared to any 'hybrid' RF MEMS tunable filter reported in current literature. Also, the half-mode substrate-integrated waveguide tunable filter is 2.5 times smaller than its full-mode counterpart while having similar performance. This dissertation also presented detailed analytical and simulation-based studies of nonlinear noise phenomena induced by Brownian motion in all-pole RF MEMS tunable filters. Two independent mathematical methods are proposed to calculate phase noise in RF MEMS tunable filters: (1) pole-perturbation approach, and (2) admittance-approach. These methods are compared to each other and to harmonic balance noise simulations using the CAD-model of the RF MEMS switch. To account for the switch nonlinearity in the mathematical methods, a nonlinear nodal analysis technique for tunable filters is also presented. In summary, it is shown that output signal-to-noise ratio degradation due to Brownian motion is maximum for low fractional bandwidth, high order and high quality factor RF MEMS tunable filters. Finally, a self-sustained microwave platform to detect the dielectric constant of organic liquids is presented in this dissertation. The main idea is to use a voltage- controlled negative-resistance oscillator whose frequency of oscillation varies according to the organic liquid under test. To make the system self-sustained, the oscillator is embedded in a frequency synthesizer system, which is then digitally interfaced to a computer for calculation of dielectric constant. Such a system has potential uses in a variety of applications in medicine, agriculture and pharmaceuticals.

碩士
國立臺灣大學
電信工程學研究所
89
In this thesis,we deal with the problem of designing FIR digital filters ,FIR and IIR filter banks with low-delay property. The design techniques based on modified primal-affine scaling algorithm and modified dual-affine scaling algorithm,in conjunc-tion with approximation schemes,are then developed for solving the resulting nonlinear optimization. For FIR digital filters,we using minimax criteria to formul-ate our design problem. With regard to the FIR and IIR digital filter banks,we using criteria to formulate our design problem,and some simulation results are provided for illustration and comparision. To compare with the L2 design in[26][27],we find the L1 design we proposed have the satisfactory design results.

碩士
國立臺灣大學
電信工程學研究所
94
Owing to the increasing use of digital signal processing, the design of digital filters and filter banks plays an important role in this filed .In this thesis we present genetic algorithm and WLS algorithm with CORDIC algorithm for QMF banks and non-uniform division filter banks based on 1-D IIR all-pass filters with discrete coefficients . The resulting two-channel QMF banks and non-uniform division filter banks can possess approximately linear phase response without magnitude distortion. The effectiveness of the proposed techniques is achieved by forming an appropriate Chebyshev approximation of a desired phase response . In each chapter，we design some examples and the simulation examples are demonstrated and satisfy，