/library/oar/handle/123456789/78457 2025-11-11T06:00:43Z 2025-11-11T06:00:43Z /library/oar/handle/123456789/101544 2022-09-13T06:28:28Z 2014-01-01T00:00:00Z

Title: Implementation of an automatic phoneme recognition system Abstract: Automatic Speech Recognition (ASR) is becoming increasingly more popularly used in most applications in today's technologies. Over the past years, a lot of research has been carried out, with the current trend being speech recognition in speaker-independent continuous speech environments with large vocabulary. However, most of the research carried out so far focused on software applications, and not much work has been carried out on the design of hardware recognisers, which is also necessary in order to attain further improvement in the field of ASR. Over the past years, ASR was not used widely in potential applications due to a number of limitations, such as processing power and limited hardware resources [1]. However, with today's advances in customised hardware, the design of ASR systems on-chip has become more feasible. In this research, different phoneme recognition systems for multi-speaker continuous speech environments are being proposed. The feature extraction stage was based on the Discrete Wavelet Transform (DWT), and for the classification stage, different Artificial Neural Networks (ANNs) and Support Vector Machines (SVMs) were analysed. From the methods considered, the One-against-one SVM method provided the highest recognition rates. Furthermore, a priorities scheme was also added, so that the three most likely phoneme representations were obtained at the output. The software implementation of this phoneme recognition system has the potential to achieve an accuracy of 75.41%, for the recognition of 42 phoneme classes from the DARPA TIMIT Acoustic-Phonetic Continuous Speech Corpus (TIMIT) corpus. The results obtained were either comparable, or slightly better, than the best results found in the literature [2] - [7], which evaluated their systems on the TIMIT corpus. On average, the potential recognition rate which can be achieved with the proposed phoneme recognition system, results in an increase in accuracy of approximately 6%, when compared to the phoneme recognition systems presented in [2] - [7]. However, the proposed system is more adequate to be implemented on hardware. The phoneme recognition system was then designed on a dedicated chip, in order to evaluate its potential into becoming a portable and efficient system which can be employed in battery-powered devices. The final design can provide a speed which is approximately 4 times faster than the software-based approach, and consumes only 12.Sm W, making it appealing to mobile devices. The performance results obtained from the hardware design demonstrated that this system is a promising basis for future hardware ASR systems. Description: PH.D.

2014-01-01T00:00:00Z /library/oar/handle/123456789/101436 2022-10-07T08:10:48Z 2010-01-01T00:00:00Z

Title: High performance CMOS frequency synthesis architectures for microwave applications Abstract: The motivation of this research was to investigate novel low voltage microwave frequency synthesizer architectures which can provide a wide range of frequencies with an acceptable phase noise response, integrated spurious level and power consumption as required by modem communication systems. In particular the work undertaken and described in this dissertation can be classified into three main parts. The first part reports the study, design and silicon implementation of a low voltage 1.6 GHz quadrature output integer-N phase locked loop based frequency synthesizer with an on-chip regulated DC-DC converter in order to achieve a wide tuning range with negligible effect on the phase noise and spurious level. The second part discusses the investigation of the use of a Micro-electro-mechanical Systems (MEMS) based tuneable inductor both as a means to extend the frequency tuning range and also to facilitate and improve the voltage-controlled oscillator design in terms of phase noise response and power consumption in comparison to a design based on conventional capacitive tuning. Finally, the last part of the dissertation discusses the investigation of the spurious tone levels in novel digitally based frequency synthesis architectures for an Ultra Wideband MB-OFDM (Multi-band Orthogonal Frequency Division Multiplexing) Alliance application which requires a wide range of frequencies. Architectures are proposed as low silicon area alternatives to state-of-the-art solutions. Description: PH.D.

2010-01-01T00:00:00Z /library/oar/handle/123456789/101141 2022-08-30T07:57:32Z 2013-01-01T00:00:00Z

Title: Fast automatic beam-based alignment of the LHC collimator jaws Abstract: The CERN Large Hadron Collider (LHC) in Geneva, Switzerland is the largest and most powerful particle accelerator ever built. With a circumference of 27 km, it is designed to collide particles in two counter-rotating beams, at a centre-of-mass energy of 14 TeV to explore the fundamental forces and constituents of matter. Due to its potentially destructive high energy particle beams, the LHC is equipped with several machine protection systems. The LHC collimation system is tasked with scattering and absorbing beam halo particles before they can quench the superconducting magnets. The 108 collimators also protect the machine from damage in the event of very fast beam losses, and shields sensitive devices in the tunnel from radiation over years of operation. Each collimator is made up of two blocks or 'jaws' of carbon, tungsten or copper material. The collimator jaws need be placed symmetrically on either side of the beam trajectory, to clean halo particles with maximum efficiency. The beam orbit and beam size need to be determined for each collimator, to be able to position the jaws Within a certain number of standard deviations (beam o-) from the beam centre. Beam-based alignment is used to determine these values at every collimator location. In the alignment procedure, each jawv is moved separately towards the beam trajectory, in 5 inn steps, until a spike appears in the signal of a Beam Loss Monitoring (BLM) detector positioned a couple of metres downstream of the collimator. A balance is required between scraping enough beam to obtain a signal, avoiding automatically triggered beam extractions (or dumps) in the event. of high beam losses, and completing the alignment in the shortest time possible to allow the LHC to produce maximum luminosity. In the 2010 LHC run, almost 30 hours were required for an alignment of all collimators, and 8 beam dumps were caused due to operator mistakes. A phased development, commissioning and usage of various algorithms in the 2011-2012 LHC runs allowed the alignment time to decrease to limit over 4 hours, with no more beam dumps. The algorithms range from automatic selection of BLM thresholds during the alignment, to BLM-based feedback loops and pattern recognition of the BLM signal spikes. The BLM-based feedback loop was also successfully used by the ALFA and TOTEM particle physics experiments in Roman Pot alignment campaigns. A Roman Pot is a detector that intercepts slightly deflected particles from head-on collisions to measure the total collision rate (cross-section). An alignment simulator was developed in MATLAB based on an empirical model of the BLM detector signal steady-state and crosstalk, as well as a beam diffusion model which allows the prediction of the characteristic BLM detector signal spike and decay. The simulator is targeted at validating possible future alignment algorithms which would otherwise require dedicated beam tests. A new collimator design for future LHC operation envisages Beam Position Monitor (BPM) pick-up buttons embedded inside the jaws. The BPM's will provide an accurate and continuous measurement of the beam centres without requiring BLM-based alignment. One quarter of the LHC collimators (tertiary collimators and IR6 secondary collimators) will be replaced with the new design, as foreseen since several years. Hence, an algorithm to automatically position the jaws around the beam centre at a large jaw gap was developed and tested with a prototype mock-up collimator installed in the Super Proton Synchrotron (SPS). Alignment times of approximately 20 s were reached. The work described in this dissertation was adopted by CERN for the first LHC running period (2008 - 2013). It will continue to be used in future operation post2015 after a two-year shutdown, in which the machine will be upgraded to be able to operate at the design parameters. Description: PhD

2013-01-01T00:00:00Z /library/oar/handle/123456789/100906 2022-08-31T06:20:46Z 2014-01-01T00:00:00Z

Title: Optimisation of the field description for the Large Hadron Collider during beam commissioning Abstract: The Large Hadron Collider (LHC) at CERN is a 27km circumference particle accelerator. It is made out of 1232 superconducting dipole magnets and 392 superconducting quadrupole magnets, which guide the particles around the LHC. It is designed to have two counter-rotating beams of particles collided at nominal centre of mass energy of 14 TeV to study and explore the fundamental forces and constituents of matter. The operation of the LHC is a challenging task. All the superconducting magnets have to be controlled simultaneously in order to successfully steer the beams around the LHC. Any field variations of the superconducting magnets, which are not immediately corrected, will affect the trajectories of the beam and therefore the LHC performance. Unfortunately, a feedback control system based solely on beam-based measurements is not enough and therefore, a feed-forward control system is also required. The feed-forward control system is used to predict the behaviour of the field variations such that it reduces the load from the feedback system. The Field Description for the Large Hadron Collider (FiDeL) is the feed-forward control system designed to predict the magnetic field variations that result from the inherent properties of the superconducting magnets. This system is based on the analysis of the data obtained from the magnetic measurements of all the superconducting magnets before the magnets were installed in the LHC. This thesis is focused on determining the precision of the FiDeL model and optimising it where necessary. However, in this work, the magnetic behaviour is not studied through magnetic measurements of individual magnets, but through beam-based measurements performed during the operation of the LHC. This is done by measuring the tune and the chromaticity of the LHC, two important beam parameters which ensure beam stability. These two parameters are directly dependent on the magnetic field quality and the control of its harmonic content. Therefore, by analysing such beam-based measurements, the behaviour of the FiDeL model is compared to the behaviour of the whole LHC. Furthermore, this study also gives an outlook of the FiDeL model at the LHC nominal operation of 7 TeV. This consists of studying the main elements of FiDeL that can become critical at such operation in order to anticipate any possible issues and how these can be overcome. The work presented in this thesis has been adopted by CERN during the first LHC running period (2008-2013). It will continue to be used in 2015, following the two- year shutdown after which the LHC will be operating at the nominal design parameters. Description: PH.D.

2014-01-01T00:00:00Z