Windows are a weakness in building facade sound transmission loss (STL). This coupled with the detrimental effects of excessive noise exposure on human health including: annoyance, sleep deprivation, hearing impairment and heart disease, is the motivation for this investigation of the STL improvements active noise control (ANC) of windows can provide. Window speaker development, ANC window experiments and analytical modelling of ANC windows were investigated. Five different window speaker constructions were characterised then compared with a previously developed window speaker. ANC window testing used three different ANC configurations and was performed in two different environments, one with a reverberant receiving room, and the other with an anechoic receiving room. Optimisation of ANC systems with particular control source locations was the aim of the modelling. This enabled comparison with the ANC window tests and would aid in further development of ANC windows. Window speaker constructions were characterised by sound pressure level (SPL) measurements performed in an anechoic room. These measurements were made in a way that enabled comparison with the previously developed window speaker. Total sound energy reduction calculations were used to determine the relative performance of the tested ANC windows. An STL model, based on a modal panel vibration model, was initially created and verified against published STL data before it was expanded to include ANC control sources. The model was used to simulate the performed anechoic environment tests and an ideal ANC case.

The demand for low power wireless transceiver implementations has been fueled by multiple applications in the recent decades, including cellular systems, wireless local area networks, personal area networks, biotelemetry and sensor networks. Dynamic range, which is set by linearity and sensitivity performance, is a critical design metric in many of these systems. Both linearity and sensitivity requirements continue to become progressively challenging in many systems due to greater spectrum usage and the need for high data rates respectively. The objective of this research is to investigate power-efficient circuit techniques for reducing the power requirement in receiver front-ends without compromising the dynamic range performance. In the first part of the dissertation, a low power receiver down-converter topology for enhancing dynamic range performance is presented. Current mode down-converters with passive mixer cores have been shown to provide excellent dynamic range performance. However, in contrast to a current commutating Gilbert cell, these down-converters require separate bias current paths for the RF transconductor and the baseband transimpedance amplifier. The proposed topology reduces the power requirement of conventional current mode passive down-converter by sharing the bias current between the transconductance and transimpedance stages. This is achieved without compromising the available voltage headroom for either stage, which is a limitation of bias-sharing based on the use of stacked stages. The dynamic range of the basic bias-current-shared topology is further enhanced through suppression of low frequency noise and IM3 products. Two variants of the down-converter, employing a broadband common-gate and a narrowband common-source input stage, are implemented in a 0.18-μm CMOS technology. The dynamic range performance of the architecture is analyzed. Finally, a prototype of a full direct-conversion receiver implementation with quadrature outputs and integrated LO synthesis is demonstrated. A power-efficient oscillator design for phase noise minimization is presented in the second part of this dissertation. This design is targeted towards multi-radio platforms where several communication links operate simultaneously over multiple frequency bands. Blockers from concurrently operating radios present a major design challenge. The blockers not only make the frontend linearity requirement more stringent but also degrade receiver sensitivity through reciprocal mixing with the phase noise sidebands of LO. Phase noise minimization is thus critical for ensuring high sensitivity in frequency bands where large blockers are present and not sufficiently attenuated by pre-select filters. A capacitive power combining technique in oscillators is introduced to improve phase noise performance. By combining this approach with current reuse, the phase noise is reduced at lower power, compared to conventional LC oscillators. This leads to improved power efficiency. Moreover, the technique mitigates modeling uncertainty arising from phase noise reduction through simultaneous impedance and current scaling. The mode selection in this oscillator, which employs multiple coupled resonators, is analyzed and the impact of coupling on far-out phase noise performance is discussed. Multi-mode oscillation can potentially arise in other oscillator topologies too, e.g., in multiphase oscillators. Mode selection in a widely used transistor-coupled quadrature oscillator is analyzed in detail in the final part of the dissertation. The analysis shows how cross-compression among multiple competing modes can lead to suppression of non-dominant modes in the steady state. / text

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / In this study the possible use of smart structures for noise control in a higher order acoustic duct was considered. The best option for this control was the use of axial splitters in the duct in order to prevent higher order mode propagation. It is possible to perform the active noise control in each splitter section by using a single channel control system. The use of smart structures takes advantage of the splitter plate and uses it as the control source, which substitutes the traditional loudspeakers used in active noise control systems. In order to evaluate the possibility of the noise control using smart structures, an analytical model of a thin plate with piezoelectric actuators was built then the acoustic field generated by this vibrating structure inside of the duct was obtained. However, to obtain the acoustic field inside an splitted duct, a numerical method such as the Component Mode Synthesis has to be used. Using the equation of the acoustic field generated in the duct by the plate, it was possible to obtain the acoustic field inside the splitted duct. After that, the active noise control simulations for harmonic excitations were performed and the influence of the size of the plate excited by the PZT actuators was studied. Finally the active control for random noise was simulated, in which the number of actuators in the plate was changed. In conclusion, it is possible to say that the smart structures can be used in active noise control of ducts with splitters and the advantages and disadvantages of the conveyed technique were presented. / Neste trabalho, foi estudada a proposta da utilização de estruturas inteligentes para o controle de ruído em um duto acústico com propagação de modos de alta ordem. A técnica mais adequada para este controle foi o particionamento do duto a fim de planificar as ondas que se propagam. Nesta região particionada, é possível realizar o controle ativo de ruído utilizando apenas um sensor e um atuador para cada lado da partição. A aplicação das estruturas inteligentes é proposta no sentido de aproveitar a placa particionadora para que, com a sua vibração, atuará como a fonte secundária necessária para o controle. Para a avaliação da possibilidade de controle utilizando esta técnica, primeiramente foi modelado o comportamento de uma placa instrumentada com atuadores piezoelétricos e, em seguida, obtida a modelagem analítica do campo sonoro gerado por uma estrutura vibrante no interior de um duto. Porém, a obtenção do campo acústico em um duto particionado não é facilmente obtido, sendo, então, realizada através da técnica de Síntese Modal de Componentes. Utilizando as equações do duto excitado por uma estrutura vibrante na técnica de síntese modal, foi possível obter campo acústico gerado no interior de um duto particionado. A partir disto, foram realizados simulações de controle ativo de ruído variando o trecho da placa a ser excitado para tons puros e para ruídos de banda estreita. Nesta última situação também foi avaliada a influência da quantidade de atuadores instalados. Concluiu-se deste trabalho que é possível a utilização de estruturas inteligentes no controle ativo de ruído em dutos particionados, sendo apresentadas suas vantagens e desvantagens. / Mestre em Engenharia Mecânica

Universidade Federal do Triângulo Mineiro / This work is dedicated to the study of the problem of active noise control, evaluating some numerical and experimental methodologies. The analysis is restricted to the case of noises in ducts, in which the acoustic propagation phenomena are modeled. Four approaches for this type of models are presented. The first one is formulated by using the basic equations of the acoustics. This procedure generates an infinite dimension model of the duct. In the second approach, the infinite model is truncated by using Taylor s series. The third approach performs a modal expansion using the poles of the infinite dimension model, and, in the fourth, it is also considered a modal expansion, but in this case, by taking into account zeros and poles of the infinite dimension model. The four models studied are discussed and compared in the present contribution. A second part of this work is concerned with active noise control techniques. Monochannel (which uses only a loud speaker and a microphone) and multi-channel (which uses several loud speakers and microphones) controllers are studied. The studied active noise controllers use LMS adaptive algorithms. The noise signals are filtered using X-LMS techniques. These types of controller are usually simple and robust. The coefficients of the controller (modeled as a digital filter) are determined by using an online adaptive procedure looking for minimizing the noise levels. The control methodologies are tested numerically by using the mathematical model of the acoustic duct proposed. With the aim of validating experimentally these controllers a test rig instrumented with loud speakers and microphones was built, and the algorithms were implemented using a personal computer. At the remaining, the numerical and experimental results are discussed and some suggestions are presented in order to continue future works. / Este trabalho formula e discorre sobre o problema de controle ativo de ruído e avalia algumas metodologias de controle tanto numérica como experimentalmente. A análise é restrita ao caso de ruídos em dutos, onde o fenômeno da propagação acústico é analiticamente modelado. Apresentam-se quatro abordagens para tal modelagem. A primeira, formulada a partir das equações fundamentais da acústica, gera um modelo de dimensão infinita para o duto. A segunda aproxima o modelo infinito por uma série truncada de Taylor. A terceira formulação realiza uma expansão modal, a partir dos pólos do modelo de dimensão infinita e a quarta, também realiza uma expansão modal, mas considera tanto os pólos como os zeros do modelo infinito dimensional. No trabalho são discutidos e comparados os quatro modelos numéricos propostos. Numa segunda parte este trabalho discorre-se sobre diversas técnicas de controle ativo de ruído em dutos. São estudados controladores do tipo mono canal, que utilizam um sensor e um atuador apenas e controladores do tipo multicanal, com vários sensores e atuadores. Todos os controladores ativos de ruído (CAR) estudados utilizam algoritmos adaptativos do tipo LMS (Least Mean Square) e técnicas de filtragem-X LMS. Este tipo de controlador tem como características marcantes à simplicidade e a robustez. Os coeficientes do controlador, modelado como um filtro digital, são adaptados on-line segundo uma estratégia que busca minimizar os ruídos não desejados. Estas metodologias de controle são testadas numericamente a partir do modelo matemático proposto para o duto acústico. Para avaliar também experimentalmente tais controladores, montou-se uma bancada de testes constituída por um duto de PVC instrumentada com alto falantes e microfones sendo os algoritmos de controle implementados em um microcomputador pessoal devidamente configurado. O trabalho encerra discutindo os resultados numéricos e experimentais obtidos e sugerindo desdobramentos a serem investigados no futuro. / Doutor em Engenharia Mecânica

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Acoustical noises are known as pollution sources that cause adverse effects in human life. Considerable investigations have been done to development of the new technologies in Active Noise Control. This work presents and experimentally analyses algorithms of Active Noise Control in Ducts presented in literature, including Feedforward algorithms, Feedback algorithms and Hybrid algorithms that uses both concepts. The identification of secondary path and feedback path is presented and solutions are discussed. In this way, methods of off-line and on-line modeling are presented. A short introduction about acoustics in ducts is presented and some effects that the acoustical noise can cause in human being are also discussed. Problems about the sensors and actuators displacement in the duct system, causality and signal conditioning are also argued here. An introduction about Digital Signal Processors (DSPs) e some particularities found in the development of this works are presented. In order to evaluate the control algorithms performance, an experimental acoustic duct using a standard PVC water pipe was built, where those algorithms were implemented in a DSP platform TMS320LF240A from Texas Instruments®. An analysis is done about the difficulty and recourses used for which algorithm implemented. This work ends presenting and discussing the obtained results for the different control procedures studied and pointing to some future works. / Ruídos acústicos são conhecidos como fontes de poluição sonora que podem causar efeitos adversos na vida humana. Para solucionar estes problemas, interesse considerável tem sido mostrado em Controle Ativo de Ruído. O intuito deste trabalho é estudar e analisar os principais algoritmos de Controle Ativo de Ruído presentes na literatura, incluindo algoritmos de malha aberta (Feedfoward) e de malha fechada (Feedback), bem como um sistema híbrido que utilize os dois conceitos. Os problemas relacionados aos caminhos secundário e de realimentação são apresentados e algumas soluções são discutidas. Neste âmbito, metodologias de modelagem off-line e online são apresentadas. Uma pequena introdução à acústica básica em dutos é apresentada e alguns efeitos que o ruído acústico pode causar ao ser humano são discutidos. Uma discussão é realizada a respeito do arranjo físico do sistema, incluindo escolha e posicionamento dos transdutores eletroacústicos. Problemas de causalidade e do condicionamento de sinais também são discutidos. Uma introdução a respeito dos Processadores Digitais de Sinais (DSPs) e algumas particularidades encontradas durante o desenvolvimento deste trabalho são apresentadas. Para validar a performance de alguns algoritmos de controle, montou-se uma bancada experimental constituída de um duto hidráulico de PVC utilizado como duto acústico, onde estes algoritmos foram implementados em linguagem C em uma plataforma DSP da Texas Instruments do tipo TMS320LF240A. Uma análise é realizada com respeito à dificuldade e recursos utilizados por cada algoritmo implementado. Resultados e discussões são apresentados com respeito à performance dos sistemas de controle. / Mestre em Engenharia Mecânica

L’atténuation des bruits gênants dans une automobile est classiquement réalisée par ajustement des caractéristiques mécaniques du véhicule : masse, raideur et amortissement. C’est une approche dite passive. Malheureusement, elle induit un ajout de masse important pour traiter les basses fréquences. Le contrôle actif de bruit (atténuation d’un bruit par superposition d’un contrebruit) est actuellement envisagé comme une solution possible à ce problème. L’objectif de cette thèse est d’évaluer les performances atteignables par cette solution. Un système acoustique étant par essence fortement résonant, sa modélisation sur une large plage de fréquence conduit à des modèles d’ordre élevé, pour l’obtention desquels une méthode d’identification appropriée doit être utilisée. C’est la méthode dessous espaces par approche fréquentielle dans le domaine continu qui a été retenue.La traduction du cahier des charges conduit à un problème de régulation multivariable H1 multi-objectif et multi-modèle avec contrainte de stabilité forte. Par ailleurs, actionneurs et capteurs ne sont pas colocalisés et on ne mesure pas la perturbation à rejeter. La volonté d’évaluer au plus près les performances atteignables justifie la résolution du problème par optimisation non lisse. Cette approche évite tout pessimisme, mais nécessite de par son caractère local une bonne initialisation et une structuration du régulateur parcimonieuse.La méthodologie proposée a été validée en simulation et expérimentalement. Elle permet une évaluation et une comparaison précises des performances atteignables en fonction des contraintes sur les mesures et les moyens d’action disponibles. / Classical methods used for noise reduction in cars are based on adjusting the mechanical properties: mass, stiffness and damping. They are qualified as passive and induce significative addition of weight for reducing low frequency noises. Active noise control is seen as a possible solution to achieve low frequency noise attenuation and weight reduction.The goal of this work is to evaluate achievable performances with such solution.Acoustic enclosures are known to be resonant systems of highorder. Obtaining a model of it, therefore requires a suitable identification method. The approach chosen is based on subspace methods. It processes data in the frequency domain for obtaining a continuous time model.The control problem derived from the specifications is a MIMO H1, multi-objective and multi-model problem with a strong stability constraint. Futhermore, actuators and sensors are not-colocated, and no measure of the disturbance is available. In order to precisely evaluate the achievable performances, this problem is solved using non smooth optimization.Such approach ensures the absence of pessimism, but requires an appropriate initialization and a parsimonious controller structure, because it does not ensure convergence toward the global optimum. The proposed methodology was validated in simulation and experimentally. It allows a precise evaluation and comparison of achievable performances according to the constraints on available measures and means of action.

The thesis deals with the analysis of technology used during the design of headphones with integrated active ambient noise cancellation and examines the possibilities of using adaptive filters to simplify development and achieve more effective attenuation.

We are seeing a revolution in computing and artificial intelligence; intelligent machines have become ingrained in and improved every aspect of our lives. Despite the increasing number of intelligent devices and breakthroughs in artificial intelligence, we have yet to achieve truly intelligent environments. Audio is one of the most common sensing and actuation modalities used in intelligent devices. In this thesis, we focus on how we can more robustly integrate audio intelligence into a wide array of resource-constrained platforms that enable more intelligent environments. We present systems and methods for adaptive audio filtering that enables us to more robustly embed acoustic intelligence into a wide range of real time and resource-constrained mobile, embedded, and cyber-physical systems that are adaptable to a wide range of different applications, environments, and scenarios. First, we introduce methods for embedding audio intelligence into wearables, like headsets and helmets, to improve pedestrian safety in urban environments by using sound to detect vehicles, localize vehicles, and alert pedestrians well in advance to give them enough time to avoid a collision. We create a segmented architecture and data processing pipeline that partitions computation between embedded front-end platform and the smartphone platform. The embedded front-end hardware platform consists of a microcontroller and commercial-off-the shelf (COTS) components embedded into a headset and samples audio from an array of four MEMS microphones. Our embedded front-end platform computes a series of spatiotemporal features used to localize vehicles: relative delay, relative power, and zero crossing rate. These features are computed in the embedded front-end headset platform and transmitted wirelessly to the smartphone platform because there is not enough bandwidth to transmit more than two channels of raw audio with low latency using standard wireless communication protocols, like Bluetooth Low-Energy. The smartphone platform runs machine learning algorithms to detect vehicles, localize vehicles, and alert pedestrians. To help reduce power consumption, we integrate an application specific integrated circuit into our embedded front-end platform and create a new localization algorithm called angle via polygonal regression (AvPR) that combines the physics of audio waves, the geometry of a microphone array, and a data driven training and calibration process that enables us to estimate the high resolution direction of the vehicle while being robust to noise resulting from movements in the microphone array as we walk the streets. Second, we explore the challenges in adapting our platforms for pedestrian safety to more general and noisier scenarios, namely construction worker safety sounds of nearby power tools and machinery that are orders of magnitude greater than that of a distant vehicle. We introduce an adaptive noise filtering architecture that allows workers to filter out construction tool sounds and reveal low-energy vehicle sounds to better detect them. Our architecture combines the strengths of both the physics of audio waves and data-driven methods to more robustly filter out construction sounds while being able to run on a resource-limited mobile and embedded platform. In our adaptive filtering architecture, we introduce and incorporate a data-driven filtering algorithm, called probabilistic template matching (PTM), that leverages pre-trained statistical models of construction tools to perform content-based filtering. We demonstrate improvements that our adaptive filtering architecture brings to our audio-based urban safety wearable in real construction site scenarios and against state-of-art audio filtering algorithms, while having a minimal impact on the power consumption and latency of the overall system. We also explore how these methods can be used to improve audio privacy and remove privacy-sensitive speech from applications that have no need to detect and analyze speech. Finally, we introduce a common selective audio filtering platform that builds upon our adaptive filtering architecture for a wide range of real-time mobile, embedded, and cyber-physical applications. Our architecture can account for a wide range of different sounds, model types, and signal representations by integrating an algorithm we present called content-informed beamforming (CIBF). CIBF combines traditional beamforming (spatial filtering using the physics of audio waves) with data driven machine learning sound detectors and models that developers may already create for their own applications to enhance and filter out specified sounds and noises. Alternatively, developers can also select sounds and models from a library we provide. We demonstrate how our selective filtering architecture can improve the detection of specific target sounds and filter out noises in a wide range of application scenarios. Additionally, through two case studies, we demonstrate how our selective filtering architecture can easily integrate into and improve the performance of real mobile and embedded applications over existing state-of-art solutions, while having minimal impact on latency and power consumption. Ultimately, this selective filtering architecture enables developers and engineers to more easily embed robust audio intelligence into common objects found around us and resource-constrained systems to create more intelligent environments.

An active noise control (ANC) system has been applied to the problem of attenuating low-frequency tonal noise inside small enclosures. The intended target application of the system was the reduction of the engine firing frequency inside heavy equipment cabins. The ANC system was based on a version of the filtered-x LMS adaptive algorithm, modified for the minimization of acoustic energy density (ED), rather than the more traditional minimization of squared acoustic pressure (SP). Three loudspeakers produced control signals within a mock cabin composed of a steel frame with plywood sides and a Plexiglas® front. An energy density sensor, capable of measuring acoustic pressure as well as acoustic particle velocity, provided the error signal to the control system. The ANC system operated on a single reference signal, which, for experiments involving recorded tractor engine noise, was derived from the engine's tachometer signal. For the low frequencies at which engine firing occurs, experiments showed that ANC systems minimizing ED and SP both provided significant attenuation of the tonal noise near the operator's head and globally throughout the small cabin. The tendency was for ED control to provide a more spatially uniform amount of reduction than SP control, especially at the higher frequencies investigated (up to 200 Hz). In dynamic measurement conditions, with a reference signal swept in frequency, the ED control often provided superior results, struggling less at frequencies for which the error sensor was near nodal regions for acoustic pressure. A single control channel often yielded performance comparable to that of two control channels, and sometimes produced superior results in dynamic tests. Tonal attenuation achieved by the ANC system was generally in excess of 20 dB and reduction in equivalent sound level for dynamic tonal noise often exceeded 4 dB at the error sensor. It was shown that temperature changes likely to be encountered in practice have little effect on the initial delay through the secondary control path, and are therefore unlikely to significantly impact ANC system stability in the event that a fixed set of system identification filter coefficients are employed.

This thesis provides analytical and experimental proof-of-concept for a new feedback-controlled sound transmission control module for use in an active segmented partition (ASP) array. The objective of such a module is to provide high transmission loss down to low audible frequencies while minimizing the overall mass of the module. This objective is accomplished in the new module by using actively controlled panels in conjunction with analog feedback controllers. The new module also overcomes two limitations that exist in current ASP modules: the inability to control broadband random-noise and the lack of bidirectional control through the module. Overcoming these limitations represents an important advancement in the research area of actively controlled partitions and broadens the number of potential applications for ASP arrays. Analogous circuit models were developed and used to predict the performance of the new ASP module under feedback control. The preliminary design consists of two loudspeaker drivers mounted back-to-back in a duct, with two decoupled analog feedback controllers connected to reduce the vibration of the loudspeaker cones. It was found that the classical analogous circuit model of a loudspeaker proved inadequate for modeling the low- and mid-frequency transmission loss due to resonance effects of the loudspeaker surround. An enhanced model of a loudspeaker was then used to account for this phenomenon and more accurately predict the transmission loss behavior. An experimental proof-of-concept module was constructed using two 10 cm diameter loudspeaker drivers, two accelerometers, and other off-the-shelf materials. The two analog feedback controllers used in the module were designed and built using measured frequency response function techniques. The passive and active transmission loss of the module was measured using a plane-wave tube. Transmission loss of broadband random-noise in excess of 50 dB was achieved between 100 Hz and 2 kHz. The experimental transmission loss results validated the numerical model and showcased the transmission loss performance of the new module design.