Parallel and distributed programming constitutes a highly promising approach to improving the performance of many applications. However, in comparison to sequential programming, many new problems arise in all phases of the development cycle of this kind of applications. For example, in the analysis phase of parallel/distributed programs, the programmer has to decompose the problem (data and/or code) to find the concurrency of the algorithm. In the design phase, the programmer has to be aware of the communication and synchronization conditions between tasks. In the implementation phase, the programmer has to learn how to use specific communication libraries and runtime environments but also to find a way of debugging programs. Finally, to obtain the best performance, the programmer has to tune the application by using monitoring tools, which collect information about the application's behavior. Tuning can be a very difficult task because it can be difficult to relate the information gathered by the monitor to the application's source code. Moreover, tuning can be even more difficult for those applications that change their behavior dynamically because, in this case, a problem might happen or not depending on the execution conditions.It can be seen that these issues require a high degree of expertise, which prevents the more widespread use of this kind of solution. One of the best ways to solve these problems would be to develop, as has been done in sequential programming, tools to support the analysis, design, coding, and tuning of parallel/distributed applications. In the particular case of performance analysis and/or tuning, it is important to note that the best way of analyzing and tuning parallel/distributed applications depends on some of their behavioral characteristics. If the application to be tuned behaves in a regular way then a static analysis (predictive or trace based) would be enough to find the application's performance bottlenecks and to indicate what should be done to overcome them. However, if the application changes its behavior from execution to execution or even dynamically changes its behavior in a single execution then the static analysis cannot offer efficient solutions for avoiding performance bottlenecks. In this case, dynamic monitoring and tuning techniques should be used instead. However, in dynamic monitoring and tuning, decisions must be taken efficiently, which means that the application's performance analysis outcome must be accurate and punctual in order to effectively tackle problems; at the same time, intrusion on the application must be minimized because the instrumentation inserted in the application in order to monitor and tune it alters its behavior and could introduce performance problems that were not there before the instrumentation. This is more difficult to achieve if there is no information about the structure and behavior of the application; therefore, blind automatic dynamic tuning approaches have limited success, whereas cooperative dynamic tuning approaches can cope with more complex problems at the cost of asking for user collaboration. We have proposed a third approach. If a programming tool, based on the use of skeletons or frameworks, has been used in the development of the application then much information about the structure and behavior of the application is available and a performance model associated to the structure of the application can be defined for use by the dynamic tuning tool. The resulting tuning tool should produce the outcome of a collaborative one while behaving like an automatic one from the point of view of the application developer.

This thesis describes and evaluates how extending Intel's Concurrent Collections (CnC) programming model can address the problem of hybrid programming with high performance and low energy consumption, while retaining the ease of use of data-flow programming. The CnC model is a declarative, dynamic light-weight task based parallel programming model and is implicitly deterministic by enforcing the single assignment rule, properties which ensure that problems are modelled in an intuitive way. CnC offers a separation of concerns by allowing algorithms to be expressed as a two stage process: first by decomposing a problem into components and specifying how components interact with each other, and second by providing an implementation for each component. By facilitating the separation between a domain expert, who can provide an accurate problem specification at a high level, and a tuning expert, who can tune the individual components for better performance, we ensure that tuning and future development, such as replacement of a subcomponent with a more efficient algorithm, become straightforward. A recent trend in mainstream desktop systems is the use of graphics processor units (GPUs) to obtain order-of-magnitude performance improvements relative to general-purpose CPUs. In addition, the use of FPGAs has seen a significant increase for applications that can take advantage of such dedicated hardware. We see that computing is evolving from using many core CPUs to ``co-processing" on the CPU, GPU and FPGA, however hybrid programming models that support the interaction between multiple heterogeneous components are not widely accessible to mainstream programmers and domain experts who have a real need for such resources. We propose a C-based implementation of the CnC model for enabling parallelism across heterogeneous processor components in a flexible way, with high resource utilization and high programmability. We use the task-parallel HabaneroC language (HC) as the platform for implementing CnC-HabaneroC (CnC-HC), a language also used to implement the computation steps in CnC-HC, for interaction with GPU or FPGA steps and which offers the desired flexibility and extensibility of interacting with any other C based language. First, we extend the CnC model with tag functions and ranges to enable automatic code generation of high level operations for inter-task communication. This improves programmability and also makes the code more analysable, opening the door for future optimizations. Secondly, we introduce a way to specify steps that are data parallel and thus are fit to execute on the GPU, and the notion of task affinity, a tuning annotation in the specification language. Affinity is used by the runtime during scheduling and can be fine-tuned based on application needs to achieve better (faster, lower power, etc.) results. Thirdly, we introduce and develop a novel, data-driven runtime for the CnC model, using HabaneroC (HC) as a base language. In addition, we also create an implementation of the previous runtime approach and conduct a study to compare the performance. Next, we expand the HabaneroC dynamic work-stealing runtime to allow cross-device stealing based on task affinity. Cross-device dynamic work-stealing is used to achieve load balancing across heterogeneous platforms for improved performance. Finally, we implement and use a series of benchmarks for testing the model in different scenarios and show that our proposed approach can yield significant performance benefits and low power usage when using a hybrid execution.

As choir masters, many organists have the responsibility of hiring and working with paid singers as well as a dedicated group of volunteer singers ranging in experience from novice to advanced. The similarities of the human voice to the pipe-organ are numerous. Using these similarities and scientific analysis of the two instruments, organists can familiarize themselves with the tuning system of the human voice. Like the pipe organ, the human voice is capable of wide variety of sounds, qualities, textures, pitches and levels of volume. Unlike an organ pipe, the voice is not a fixed resonator. The voice is the most flexible of all musical instruments. Instructing an ensemble of singers to shape their sound simultaneously is the beginning of “tuning your choral pipes.” It will be important to establish terminology with your singers in order to successfully communicate with them despite their varying levels of ability and pronunciation differences. Becoming familiar with the mechanics of the voice and an alphabet of pure vowel sounds can help organist-choir masters achieve a greater degree of success when working with singers. The stops, pipes and expression pedal of the human voice are defined by the laryngeal muscles as they relate to registration, the vocal tract shape as defined by the vowel, and the amount of volume created by the air pressure. This guide for organists covers these topics and contains exercises for the reader to apply during choral rehearsals.

This is a thesis report done as part of the Master of Science in Electronics Design Engineering given at Linköping University, Campus Norrköping. The thesis work is done at Ericsson AB in the spring of 2005. The thesis describes a method of removing variations in the tuning sensitivity of voltage-controlled crystal oscillators due to different manufacturing processes. These variations results in unwanted variations in the modulation bandwidth of the phase-locked loop the oscillator is used in. Through examination of the theory of phase-locked loops it is found that the bandwidth of the loop is dependent on the tuning sensitivity of the oscillator. A method of correcting the oscillator-sensitivity by amplifying or attenuating the control-voltage of the oscillator is developed. The size of the correction depends on the difference in oscillator-sensitivity compared to that of an ideal oscillator. This error is measured and the correct correction constant calculated. To facilitate the measurements and correction extra circuits are developed and inserted in the loop. The circuits are both analog and digital. The analog circuits are mounted on an extra circuit board and the digital circuits are implemented in VHDL in an external FPGA. Tests and theoretical calculations show that the method is valid and able to correct both positive and negative variations in oscillator-sensitivity of up to a factor ±2.5 times. The bandwidth of the loop can be adjusted between 2 to 15 Hz (up to ±8 dB, relative an unmodified loop).

Multi-band and multi-mode radios are becoming prevalent and necessary in order to provide optimal data rates across a network with a diverse and spotty landscape of coverage areas (3G, HSPA, LTE, etc.). As the number of required bands and modes increases, the aggregate cost of discrete RF signal chains justi es the adoption of tunable solutions. Tunable fi lters are one of the pieces crucial to signal chain amalgamation. The main requirements for a tunable fi lter are high unloaded quality factor, wide tuning range, high tuning speed, high linearity, and small size. MEMS technology is the most promising in terms of tuning range, quality factor, linearity and size. In addition, a fi lter that maintains a constant passband bandwidth as the center frequency is tuned is preferred since the analog baseband processing circuitry tends to be tailored for a particular signal bandwidth. In this work, a novel design technique for tunable fi lters with controlled and predictable bandwidth variation is presented. The design technique is presented alongside an analysis and modeling method for predicting the final filter response during design optimization. The method is based on the well known coupling matrix model. In order to demonstrate the design and modeling technique, a novel coupling structure for stripline fi lters is presented that results in substantial improvements in coupling bandwidth variation over an octave tuning range when compared to combline and interdigitated coupled line fi lters. In order for a coupled resonator filter to produce an equal ripple Chebyshev response, each resonator of the fi lter must be tuned to precisely the same resonant frequency. Production tuned fi lters are routinely tuned in the lab and production environments by skilled technicians in order to compensate for manufacturing tolerances. However, integrated tunable filters cannot be tuned by traditional means since they are integrated into systems on circuit boards or inside front end modules. A fixed tuning table for all manufactured modules is inadequate since the required tuning accuracy exceeds the tolerance of the tuning elements. In this work, we develop tuning techniques for the automatic in-circuit tuning of tunable filters using scalar transmission measurement. The scalar transmission based techniques obviate the use of directional couplers. Techniques based on both swept and single frequency scalar transmission measurement are developed. The swept frequency technique, based on the Hilbert transform derived relative groupdelay, tunes both couplings and resonant frequencies while the single frequency technique only tunes the center frequency. High performance filters necessitate high resonator quality factors. Although fi lters are traditionally treated as passive devices, tunable fi lters need to be treated as active devices. Tuning elements invariably introduce non-linearities that limit the useful power handling of the tunable fi lter. RF MEMS devices have been a topic of intense research for many years for their promising characteristics of high quality factor and high power handling. Control and reliability issues have resulted in a shift from continuously tunable devices to discretely switched devices. However, fi lter tuning applications require fine resolution and therefore many bits for digital capacitor banks. An analog/digital hybrid tuning approach would enable the tuning range of a switched capacitor bank to be combined with the tuning resolution of an analog tunable capacitor. In this work, a device-level position control mechanism is proposed for piezoresistive feedback of device capacitance over the device's tuning range. It is shown that piezoresistve position control is ef ective at improving capacitance uncertainty in a CMOS integrated RF MEMS variable capacitor.

<p>Since the Renaissance, the normative approach to a philosophy of music has concerned itself primarily with the subjective experience of the listener. This was not always the case. From Greek Antiquity to the Renaissance, music was considered a rigorous, mathematical discipline that shed light on objective truths concerning cosmology and cosmogony. <i>Musica theorica</i>, therefore, took precedence over musica practica and was taken much more seiously in musical scholarship. Although tension had always existed between <i>Musica theorica</i> and <i>Musica practica</i>, such tension reached its peak during the Renaissance and as a result, a shift occurred: <i>Musica theorica</i> was pushed into the background, and <i>Musica practica</i> stepped forward.</p><p>The intention of this thesis is to convince its reader of the need to bring <i>Musica theorica</i> back to its proper place in musical and philosphical scholarship, where objective answers can be found, and music's innate eternal nature is revealed. The thesis begins with a historical survey of musical scholarship that eventually brings the reader to the center of the controversy that ensued during the Renaissance, and then forward to present day discussions in philosphy of music that are concerned with music's subjective and temporal properties. It is hoped that the reader will see the need for a new shift to occur in philosophy of music that focuses on music's objective and eternal properies (that are wholly distinct from the subject experiencing it), and will come away with a new perspective regarding the interdisciplinary nature of philosphy and music.

This thesis has described a wide tuning range transconductor combining source degeneration, cross-coupled, translinear loop to achieve high linearity. The transconductance tuning range from 220£gs to 1050£gs with 1V input range and the total harmonic distortion is -50dB with 0.6Vpp input voltage. And its application to a fifth-order elliptic low-pass Gm-C filter for the front-end RF circuit is presented. In order to transform the passive element circuit into a Gm-C based filter, a GIC flow method has been used. The proposal Gm-C based filter achieve a with performance a low frequency filtering range from 5Mhz to 10Mhz by transconductance tuning.

ABSTRACT In this dissertation, several adaptive control design schemes for a class of nonlinear systems are proposed. The first topic of the research is concerned with self-tuning PID controller design. The main problem of designing PID controller is how to determine the values of three control gains, i.e., proportional gain , integral gain , and derivative gain . We attempt to use the technique of adaptive control based on the Lyapunov approach to design the PID controller for some class of partially known nonlinear systems. Three PID control gains are adjusted on-line such that better output performance can be achieved. The stability of the closed-loop PID control systems is analyzed and guaranteed by introducing a supervisory control and a modified adaptation law with projection. Second, two kinds of adaptive neural control systems including the direct and indirect neural controls are considered by using simple single auto-tuning neuron. We will first propose a novel neuron called auto-tuning neuron and use it to take place of the roles of the traditional neural networks used in the direct and indirect adaptive neural control systems. This can greatly reduce the computational time and network complexities due to the simple configuration of the auto-tuning neuron. It is also easy for hardware implementation. Third, based on the idea borrowed from natural evolution, genetic algorithm can search for optimal or near-optimal solutions for an optimization problem over the search domain. An optimization technique of real-coded genetic algorithm is used to design the PID controller by minimizing the performance index of integrated absolute error. The improvements of our results over that using other methods are also illustrated. In the last part of each section, some computer simulation results will also be provided to illustrate our proposed methods.

A novel on chip automatic tuning circuit for Gm ¡V C continuous time filter is presented. The circuit is composed of an integrator, a frequency tuning circuit, and a Q tuning circuit. A 4th order Chebyshev low pass filter is also designed with the tuning circuitry. All circuits are designed by using the parameters of TSMC 0.25um process. The power supplies are ¡Ó2.5V, and the cutoff frequency is 10MHz. The main LPF exhibits passband ripple below 4dB, and stopband attenuation over 70dB. The equilibrium time for tuning circuits is less than 3£gseconds.

Since the plus fraction of reservoir fluids has some uncertainty in its molecular weight and critical properties, equation-of-state, EOS, are generally not predictive without tuning its parameters to match experimental data. Tuning of the EOS is found to be the best method for improving the predictions of compositional reservoir simulators. The proposed strategy for tuning EOS consists of seven steps: (1) split the laboratory plus fraction to single carbon number groups, SCN, usually up to SCN 44; the last component will be C45+, (2) use set of correlations to calculate the critical properties and acentric factor for each SCN group, (3) match the saturation pressure at reservoir temperature by altering the measured value of the molecular weight of the plus fraction using the extended composition, (4) group SCN groups to multiple carbon number groups, MCN, (5) assign critical properties and acentric factor for each MCN group, (6) rematch the saturation pressure at reservoir temperature using the grouped composition, and (7) match the volumetric data by regressing on volume shift parameters of all components in grouped composition. This research shows an accurate method to split the plus fraction to SCN groups. The most accurate set of correlations to calculate the critical properties and acentric factor for each SCN group that will result in a small adjustment for the molecular weight of the plus fraction when saturation pressure is matched using the extended composition. The proposed strategy groups the extended composition to eight pseudocomponents. The binary interaction coefficients between hydrocarbons and between hydrocarbons and non-hydrocarbons are set to zero which dramatically reduces the simulation time. The strategy proposed in this research for tuning EOS to match experimental data has been tested for a wide range of C7+ mole% (4 – 25) which covers gas condensate and volatile oil samples. Also, using this strategy to tune EOS at reservoir temperature will accurately predict the fluid properties at separator conditions and saturation pressures at different temperatures. The scope of this research is to come up with an accurate and systematic technique for tuning an EOS for use in compositional simulation.