Following the trend in portable wireless communications, this dissertation explores new approaches to designing of power-critical building blocks in the elementary circuit level. Specifically, the work focuses on designs of baseband continuous-time Gm-C filter, LC-resonator quadrature oscillators, transistor-only quadrature oscillators and LC-resonator frequency dividers. The established circuits share a common design objective of low-power and low-voltage operation, where the simplicity of the demonstrated topologies serves as a basis. The dissertation is separated in two parts. The first part is dedicated for the baseband section where a 3rd-order Bessel filter is designed and fabricated. The filter comprises a set of linear transconductors, which each one is based on the operation of triode-biased transistors. According to the operation in this region and to the simplicity of the transconductor, high dynamic range can be achieved for a supply voltage as low as 1.2 V. In the other part, attempts in reducing the power consumption of two critical building blocks in a frequency synthesizer, namely, the quadrature oscillator and the first-stage frequency divider, are introduced. For the oscillators, two quadrature oscillators based on LC resonators are presented, in conjunction with a transistor-only quadrature oscillator. Quadrature signal generations in these designs are achieved by making use of the principles of ring oscillator and coupled oscillator. The last building block that is designed in this part is the LC-based injection-locked frequency divider. The single-ended Colpitts oscillator topology is used as the core circuit of the divider due to its simplicity and low-voltage property. Detailed analysis concerning the phase relationship between the input and the output leads to the implementations of differential and quadrature divider configurations. Although a silicon integration is done only for the baseband filter, the concept, the operation and the theories developed for the quadrature oscillators and the frequency dividers have been verified against simulations and measurements employing low-frequency discrete prototypes. As they are illustrated in each chapter, the established theories match closely to the measurements.
A novel frequency synthesizer with a strong emphasis on low-power consumption (2mW) was developed for this thesis. A BAW-resonator was used for the design of the high-frequencyoscillator. The BAW's high Q-Factor ensured a minimal power consumption, while providing outstanding phase noise performance. This type of frequency synthesizer can be implemented in practically every mobile wireless system standard, which is relying on batteries as the power supply, such as Bluetooth, Zigbee and others. For this work, a Bluetooth-compatible standard was taken in order to derive the system specifications. The frequency synthesizer then was designed and implemented in a 180nm CMOS process. The approach of the "Sliding-IF" architecture was chosen and modified in order to take full advantage on the BAW oscillator. Therefore a two-stage frequency translation of the signal was necessary, similar to the super-heterodyne transceiver topology. The channel selection and the frequency error correction is done at the second stage, since the BAW-based oscillator does practically not provide any frequency tuning. For this, a Delta-Sigma-Modulator based Fractional-N PLL was developed. A system supply voltage of 1.2V was chosen in order to keep power consumption as low as possible. This supply voltage can also be easily supplied by AA or AAA-batteries and a low-voltage regulator. The most important technique to reduce overall power consumption was to reduce the high-frequency requirements to the system blocks. This was done by modifying the "Sliding-IF" architecture and by the use of the BAW resonator. The minimization of the power consumption required careful choice and design of each of the system blocks. For this, several approaches were analyzed, especially the topology of each block and its optimization were important. A special emphasis on the analysis, design and discussion was also made on the noise, and especially the phase noise. A ring oscillator, a LC-VCO and a BAW-oscillator were implemented and measured in order to validate the theory and the simulation results. The complete frequency synthesizer with an Automatic Amplitude Control, Frequency Dividers, Mixer, Charge-Pump, Phase-Frequency Detector and a Delta-Sigma-Modulator was designed and implemented. This thesis was elaborated in the frame of the MiNAMI and MiMOSA projects of the European research programme (FP6).
The LHCb detector is one of the four experimental setups built to detect high-energy proton collisions to be produced by the Large Hadron Collider (LHC). Located at CERN (Geneva, Switzerland), the LHC machine and the LHCb experiment are expected to start in 2008, and will then operate for several years. Being the largest collider of its kind, the LHC will open the way to new investigations, in the very-high energies, but also in terms of statistics for the study of rare-phenomena and flavor physics. In this framework LHCb is dedicated to precise measurements of CP-violating and rare decays of beauty hadrons, in order to test (or over-constrain) the Standard Model of particle physics. From the hardware point of view, the construction of such detectors represents several challenges; one of them is the routing at a very high frequency of many signals in a harsh radiation environment. We designed to this purpose a hardware setup and a software filter which together reduce the cross-talk present in the readout of the LHCb vertex detector to a level of (1 ± 2)%, leading to an improved signal quality in the acquisition chain. From the physics point of view, many of the CP-violation measurements performed at LHCb using Bs decays, for example using Bs → Ds∓ K± decays, will require as input the Bs–Bs oscillationfrequency Δms. Thus the measurement of Bs–Bs oscillations, which are best observed using the flavor-specific Bs → Ds- π+ decays, will play an important role. We have developed a complete selection of Bs → Ds- π+ and Bs → Ds∓ K± events, based on Monte Carlo simulations. Assuming 2 fb-1 of data, we expect a Bs → Ds- π+ signal yield, after the first level of trigger, of 155 k events over a background between 0.6 k and 7.8 k events at 90% confidence level. Moreover, we assess with fast Monte Carlo studies the corresponding statistical sensitivity on the Bs oscillationfrequency, σ(Δms) = 0.008 ps-1, on the wrong tag fraction, σ(ω) = 0.003, as well as on other parameters related to the Bs-meson system. We also addressed an important aspect of the systematics associated with the Δms measurement, and developed a method to calibrate and assess the length scale. This calibration is performed through the reconstruction of secondary interactions occurring in the material of the vertex detector. We show that the statistical relative precision of this approach quickly matches 6 × 10-5, obtained from the survey measurements of the detector.
Legged robots have gained an increased attention these past decades since they offer a promising technology for many applications in unstructured environments where the use of wheeled robots is clearly limited. Such applications include exploration and rescue tasks where human intervention is difficult (e.g. after a natural disaster) or impossible (e.g. on radioactive sites) and the emerging domain of assistive robotics where robots should be able to meaningfully and efficiently interact with humans in their environment (e.g. climbing stairs). Moreover the technology developed for walking machines can help designing new rehabilitation devices for disabled persons such as active prostheses. However the control of agile legged locomotion is a challenging problem that is not yet solved in a satisfactory manner. By taking inspiration from the neural control of locomotion in animals, we develop in this thesis controllers for legged locomotion. These controllers are based on the concept of Central Pattern Generators (CPGs), which are neural networks located in the spine of vertebrates that generate the rhythmic patterns that control locomotion. The use of a strong mathematical framework, namely dynamical systems theory, allows one to build general design methodologies for such controllers. The original contributions of this thesis are organized along three main axes. The first one is a work on biological locomotion and more specifically on crawling human infants. Comparisons of the detailed kinematics and gait pattern of crawling infants with those of other quadruped mammals show many similarities. This is quite surprising since infant morphology is not well suited for quadruped locomotion. In a second part, we use some of these findings as an inspiration for the design of our locomotion controllers. We try to provide a systematic design methodology for CPGs. Specifically we design an oscillator to independently control the swing and stance durations during locomotion, then using insights from dynamical systems theory we construct generic networks supporting different gaits and finally we integrate sensory feedback in the system. Experiments on three different simulated quadruped robots show the effectiveness of the approach. The third axis of research focus on dynamical systems theory and more specifically on the development of an adaptive mechanism for oscillators such that they can learn the frequency of any periodic signal. Interestingly this mechanism is generic enough to work with a large class of oscillators. Extensive mathematical analysis are provided in order to understand the fundamental properties of this mechanism. Then an extension to pools of adaptive frequencyoscillators with a negative feedback loop is used to build programmable CPGs (i.e. CPGs that can encode any periodic pattern as a structurally stable limit cycle). We use the system to control the locomotion of a humanoid robot. We also show applications of this system to signal processing.
Amplitude and frequency are the two primary features of one-dimensional signals, and thus both are widely utilized to analysis data in numerous fields. While amplitude can be examined directly, frequency requires more elaborate approaches, except in the simplest cases. Consequently, a large number of techniques have been proposed over the years to retrieve information about frequency. The most famous method is probably power spectral density estimation. However, this approach is limited to stationary signals since the temporal information is lost. Time-frequency approaches were developed to tackle the problem of frequency estimation in non-stationary data. Although they can estimate the power of a signal in a given time interval and in a given frequency band, these tools have two drawbacks that make them less valuable in certain situations. First, due to their interdependent time and frequency resolutions, improving the accuracy in one domain means decreasing it in the other one. Second, it is difficult to use this kind of approach to estimate the instantaneous frequency of a specific oscillatory component. A solution to these two limitations is provided by adaptive frequency tracking algorithms. Typically, these algorithms use a time-varying filter (a band-pass or notch filter in most cases) to extract an oscillation, and an adaptive mechanism to estimate its instantaneous frequency. The main objective of the first part of the present thesis is to develop such a scheme for adaptive frequency tracking, the single frequency tracker. This algorithm compares favorably with existing methods for frequency tracking in terms of bias, variance and convergence speed. The most distinguishing feature of this adaptive algorithm is that it maximizes the oscillatory behavior at its output. Furthermore, due to its specific time-varying band-pass filter, it does not introduce any distortion in the extracted component. This scheme is also extended to tackle certain situations, namely the presence of several oscillations in a single signal, the related issue of harmonic components, and the availability of more than one signal with the oscillation of interest. The first extension is aimed at tracking several components simultaneously. The basic idea is to use one tracker to estimate the instantaneous frequency of each oscillation. The second extension uses the additional information contained in several signals to achieve better overall performance. Specifically, it computes separately instantaneous frequency estimates for all available signals which are then combined with weights minimizing the estimation variance. The third extension, which is based on an idea similar to the first one and uses the same weighting procedure as the second one, takes into account the harmonic structure of a signal to improve the estimation performance. A non-causal iterative method for offline processing is also developed in order to enhance an initial frequency trajectory by using future information in addition to past information. Like the single frequency tracker, this method aims at maximizing the oscillatory behavior at the output. Any approach can be used to obtain the initial trajectory. In the second part of this dissertation, the schemes for adaptive frequency tracking developed in the first part are applied to electroencephalographic and electrcardiographic data. In a first study, the single frequency tracker is used to analyze interactions between neuronal oscillations in different frequency bands, known as cross-frequency couplings, during a visual evoked potential experiment with illusory contour stimuli. With this adaptive approach ensuring that meaningful phase information is extracted, the differences in coupling strength between stimuli with and without illusory contours are more clearly highlighted than with traditional methods based on predefined filter-banks. In addition, the adaptive scheme leads to the detection of differences in instantaneous frequency. In a second study, two organization measures are derived from the harmonic extension. They are based on the power repartition in the frequency domain for the first one and on the phase relation between harmonic components for the second one. These measures, computed from the surface electrocardiogram, are shown to help predicting the outcome of catheter ablation of persistent atrial fibrillation. The proposed adaptive frequency tracking schemes are also applied to signals recorded in the field of sport sciences in order to illustrate their potential uses. To summarize, the present thesis introduces several algorithms for adaptive frequency tracking. These algorithms are presented in full detail and they are then applied to practical situations. In particular, they are shown to improve the detection of coupling mechanisms in brain activity and to provide relevant organization measures for atrial fibrillation.
Methods based on the electron spin resonance (ESR) phenomenon are non-invasive tools adopted to investigate paramagnetic systems at temperatures ranging from above 1000 K to below 1 K. Since 2008, the group of Dr. Boero has been working on a detection technique based on the integration of ESR sensors on single chips. The proposed methodology allowed to study samples in the nanoliter scale and reach a spin sensitivity at least two orders of magnitude better than the best commercially available spectrometers. The detection principle can be summarized as follows. An ESR sensitive sample is placed in close proximity to the planar inductor of an LC oscillator operating at microwave frequency. In presence of a suitable static magnetic field, the ESR phenomenon takes place. It causes a variation in the sample magnetization which translates to a variation of the inductance, leading to both a frequency shift of the oscillator (frequency detection) and a variation of the oscillation amplitude (amplitude detection). Consequently, the ESR phenomenon may be detected by tracking the operating point of the oscillator. In this thesis, I investigate the application of the aforementioned detection principle in the range from 400 MHz to 360 GHz. Firstly, a semi-integrated solution operating from 400 MHz to 610 MHz is developed for an industrial application (CTI project). In such context, the originality of the work stands in the implementation of a completely standalone portable scanner for contactless inspection which may also be used for ferromagnetic (FMR) applications and zero-field measurements. Secondly, a set of single-chip ESR detectors working from 10 GHz to 146 GHz and based on CMOS technologies are characterized from 300 K down to 10 K. Here, an ESR experiment at a frequency as high as 360 GHz can be performed thanks to the fourth harmonic signal generated by a 90 GHz detector. Conversely, the 10 GHz detector shows the best noise performance and allows to achieve the record distance resolution of 0.3 pm when used as a proximity sensor. After that, the possibility of using integrated technologies based on HEMTs is investigated so as to overcome the main limitations of the CMOS based detectors: (1) the high power consumption which denies their use below 10 K and (2) the saturation issue due to the magnitude of the intrinsic microwave magnetic field produced by the oscillators. In this context, two HEMT oscillators working at 11 GHz and 25 GHz are realized. In particular, the former achieves the record minimum power consumption (90 uW at 300 K and 4 uW below 30 K) currently reported in the literature for oscillators working in the same frequency range. Also, the proposed sensor achieves a minimum microwave magnetic field of less than 1 uT at 300 K and less than 0.1 uT below 30 K, i.e., orders of magnitude below the values achieved with previous CMOS detectors. Furthermore, an analytical model is carried out in order to estimate the minimum achievable power consumption for an LC single-ended Colpitts oscillator based on any single FET. Lastly, the DC characterization of a standing alone HEMT transistor is provided from 300 K down to 1.4 K, ranging from the standard Ids-Vs-Vds curves to the extraction of both the number of carriers and their effective mobility. The former comes from the analysis of the Shubnikov-de-Haas oscillations whereas the latter is calculated by means of Hall-effect based experiments.
Mechanical oscillators are among the most important scientific tools in the modern physics. From the pioneering experiments in 18th by founding fathers of modern physics such as Newton, Hooke and Cavendish to the ground braking experiments in the 21th century where the merge of two massive black holes 1.3 billion light-year away detected on earth by a gravitational wave detectors, the high Q mechanical oscillators were at the core of many monumental experiments in physics. Their ability to couple to many different physical quantities such as mass, charge, acceleration, electro-magnetic forces and optical fields makes them an ideal candidate for sensing applications. In addition, their intrinsically low dissipation rates (¿m) results in reduced coupling to the thermal bath. Since the invention of micro/nano-technology in the second half of the 20th century and ability to control the dimensions at micro and nano-scales, new horizon was opened up for mirco/nanomechanical oscillators. Miniaturization of the mechanical oscillators made them small and stiff enough to be used in our handheld electronics where dozens of mechanical sensors such as accelerometers and gyroscopes are used in our laptops and smartphones everyday. Besides these technological advancements, since the beginning of 21th century, a new opportunity for mechanical oscillators emerged: the idea of ¿putting mechanics into quantum mechanics¿ and observing the quantum effects of these massive classical oscillators. Aside from the numerous technical challenges for achieving this goal, two fundamental obstacles has to solved: I) Even the smallest nano-mechanical oscillators still consist of billions of atoms and molecules and are orders of magnitude more massive that the traditional ¿quantum objects¿ such as atoms and molecules. Larger mass results in smaller zero point motion¿the length scale where quantum effects are visible¿which means in order to ¿see¿ these quantum effects, we have to detect smaller displacement than ever before. II) The second challenge is the low frequency of the mechanical oscillators which makes their thermal Brownian energy, orders of magnitude larger than the quantum ground state of the oscillator¿the energy scale where the quantum effects are visible¿as n_th = kBT/h¿>>1 even for a ¿/2¿ ~ 1GHz oscillator at room temperature. Both of these obstacles, can be seen as the competition between few fundamental rates: thermal decoherence and measurement rate/mechanical frequency. Thermal decoherence is the rate at which the mechanical oscillator exchange phonons ¿ quanta of mechanical energy (h¿)¿ with its thermal environment and is given by ¿decoherence = ¯ n_th*¿m. The first obstacle translates to having the measurement rate being faster than the decoherence rate of the mechanical oscillator, ¿measurement>¿decoherence. This means in order to see the quantum coherent motion of the mechanical oscillator, we have to ¿look¿ at it before it has time to exchange random thermal energy with its environment. In other words, the life time of the quantum states of macroscopic objects are limited by their thermal decoherence rate and we have to interact with the oscillator in this short lifetime. The second obstacle on the other hand, reduces to having the mechanical frequency larger than its thermal decoherence rate, ¿m >¿decoherence. Over the past 15 years, the field of cavity opto-mechanics was very successful in improving the measurement schemes and designing ...
Similarly to mechanical structures, stable flows can exhibit resonance when perturbed by an impulsive or harmonic forcing. Swirling wakes and sloshing waves belong to this kind of flows and manifest large energy response when excited close to their natural frequencies. Although these frequencies can be predicted by linear modal analysis, the full flow dynamics differs from the modal one because entailed by the mutual cooperation of the natural modes (non-normal effects) and dependent on the oscillation amplitude (nonlinear effects). In this thesis, the response of swirling wakes subjected to a harmonic forcing is studied numerically and theoretically. Direct numerical simulations show that a large variety of helical modes can be excited and amplified in trailing vortices when a harmonic inlet or volume forcing is imposed, with the appearance of higher wavenumber modes at higher frequency. The mode-selection mechanism is shown to be directly connected to the local stability properties of the flow, and is simultaneously investigated by a WKB approximation, in the framework of weakly non-parallel flows, and by the global resolvent approach. This analysis is then extended to the case of turbulent swirling flows to investigate the physical origin of the meandering oscillations of the hub vortex, that is observed in wind turbine wakes experiments. We show as this low frequency spectral component is the result of a convectively unstable single-helix structure that oscillates at a frequency equal to one third the rotational frequency of the wind turbine rotor. Consequently, an adjoint-based technique for the passive control of these helical instabilities is proposed. We then turn our attention towards the transient decay of sloshing waves affected by a viscous friction at the containerâs wall, that exhibits a sublinear dependence in the interface velocity, i.e. a power law with an exponent smaller than one. This capillary effect is exacerbated in our experiment by placing a thin layer of foam on the liquid phase that act as a collection of air-liquid interfaces. In contrast to classical theory, we uncover the existence of a finite-time singularity in our system yielding the arrest of the sloshing oscillations in a finite time and we propose a minimal theoretical framework to capture this effect. Using first principles, we then study the effect of contact angle hysteresis on sloshing waves. We show asymptotically that, in contrast to viscous damping where the wave motion decays exponentially, the contact angle hysteresis acts as Coulomb solid friction yielding the damping rate induced by the motion of the liquid meniscus to increase at small amplitude, consistently with the experimental observation.
The aim of this principally experimental study is to understand from fluid mechanic principles why an insignificant anesthetic dose administered as a short bolus into the cerebrospinal fluid inside the subarachnoid space provides greater pain relief than a larger dose continuously injected over a longer period. The subarachnoid space is modeled as an annular gap of constant or slowly varying cross section into which a catheter is introduced. The cerebrospinal fluid is replaced by water of 37°C which has very similar properties. This fluid in the annular gap is subjected to oscillations of amplitude and frequency (heart frequency) typically found in the subarachnoid space. The anesthetic is replaced by a fluorescent dye injected through the catheter. To study its dispersion, we have developed a 400 Hz laser scanning setup with which we perform quasi-instantaneous, quantitative 3D laser induced fluorescence (LIF) as well as 2D particle image velocimetry (PIV). The experiments are supplemented by an analytical axi-symmetric model as well as an exploratory numerical model to help interpret the results. The study has identified steady streaming (a nonlinear effect associated with the fluid oscillation) and enhanced diffusion (an effect associated with oscillating shear flow) as the principal agents of dye (anesthetic) dispersion. Besides the slowly varying cross section, the catheter tip has been identified as an important cause for steady streaming. In an attempt to identify optimal injection parameters of use for clinicians, a rough parametric model of the primary factors influencing drug spread (fluid oscillationfrequency and amplitude, geometry, and injection rate) has been constructed.
Harmonic oscillators might be one of the most fundamental entities described by physics. Yet they stay relevant in recent research. The topological properties associated with exceptional points that can occur when two modes interact have generated much interest in recent years. Harmonic oscillators are also at the heart of new quantum technological applications: the long lifetime of high-Q resonators make them advantageous as quantum memories, and they are employed for narrowband processing of quantum signals, as in Josephson parametric amplifiers. The goal of this thesis is to explore different fundamental regimes of two coupled harmonic oscillators using cavity optomechanics as the experimen- tal platform. With consistent progress in attaining ever increasing Q factors, mechanical and electromagnetic resonators realize near-ideal harmonic oscillators. By parametrically modulating the nonlinear optomechanical interaction between them, an effective linear coupling is achieved, which is tunable in strength and in the relative frequencies of the two modes. Thus cavity optomechanics provides a framework with excellent control over system parameters for the study of two coupled harmonic modes. The specific optomechanical implementation employed are superconducting circuits with the vibrating top plate of a capacitor as the mechanical element. Multimode optomechanical circuits are realized, with two microwave modes interacting with one or two mechanical oscillators. The supplementary modes serve either as intermediaries in the relation of the two modes of interest, or as auxiliaries used to control a parameter of the system. Three main experimental results are achieved. First, an auxiliary microwave mode allows the engineering of the effective dissipation rate of a mechanical oscillator. The latter then acts as a reservoir for the main microwave mode with which it interacts. The microwave mode susceptibility can be tuned, resulting in an instability akin to that of a maser and in resonant amplification of incoming microwave signals with an added noise close to the quantum minimum. Second, we study the conditions for a nonreciprocal interaction between two microwave modes, when the information flows in one direction but not in the other. The two modes interact through two mechanical oscillators, leading to frequency conversion between the two cavities. Dissipation in the mechanical modes is essential to the scheme in two ways: it provides a reciprocal phase necessary for the interference and eliminates the unwanted signals. Third, level attraction between a microwave and a mechanical mode is demonstrated, where the eigenfrequencies of the system are drawn closer as the result of interaction, rather distancing themselves as in the more usual case of level repulsion. The phenomenon is theoretically connected to exceptional points, and a general classification of the possible regimes of interaction between two harmonic modes is exposed, including level repulsion and attraction as special cases.