No description available.

Made available in DSpace on 2014-06-12T15:55:53Z (GMT). No. of bitstreams: 2 arquivo2330_1.pdf: 3127189 bytes, checksum: 01d7c49fae931bcb459bfda7372af32e (MD5) license.txt: 1748 bytes, checksum: 8a4605be74aa9ea9d79846c1fba20a33 (MD5) Previous issue date: 2009 / Conselho Nacional de Desenvolvimento Científico e Tecnológico / A solução de problemas complexos em várias áreas do conhecimento humano, tais como: análise de investimento no setor bancário, análise e visualização de imagens médicas em tempo real, indústria de óleo e gás, etc. que utilizam muitas vezes algoritmos complexos e/ou uma grande massa de dados, têm requerido cada vez mais sistemas computacionais de alto desempenho para seu processamento. Estes aplicativos, em sua maioria, devido a sua grande massa de dados, grandes laços de processamento em seus procedimentos, podem consumir dias ou até meses de trabalho, em computadores de processamento seqüencial, para apresentar o resultado final. Existem casos em que este tempo excessivo pode inviabilizar um projeto em questão, por perder o time to market de um produto. Diferentes tecnologias e estruturas de dados têm sido sugeridas para lidar com tais problemas, visando uma melhor customização, tentando retirar o melhor da arquitetura e do sistema, seja em termos de software como de hardware. Dentre estas arquiteturas hw/sw, optamos neste trabalho ao estudo de uma solução baseada em FPGAs (Field Programmable Gate Arrays) como um co-processador. O uso deste dispositivo permite uma nova abordagem do problema. Agora, um determinado aplicativo poderia ser particionado em duas partes: a primeira, aquela com características de controle, processo seqüencial, continuaria sendo executado no host com processadores genéricos, enquanto que a parte com os grandes laços de processamento seriam processados, com maior desempenho por explorar o paralelismo, nos co-processadores com FPGAs. Porém, a movimentação dos dados entre a memória principal do host e a memória externa do FPGA é considerada um grande gargalo para o processamento em hardware. Vários autores em seus trabalhos demonstram o desempenho alcançado com o uso de processamento em hardware, mas consideram que os dados já estão na memória externa do FPGA. Poucos comentam sobre a perda de desempenho quando se considera a movimentação de dados. Neste trabalho foram estudadas técnicas de particionamento de grandes matrizes densas, reuso de dados e as estratégias que melhor se adéquam para algumas arquiteturas estudadas neste trabalho. As latências desta movimentação de dados entre o host e o co-processador em FPGA foram o foco deste trabalho também. Concluímos com um estudo de caso onde propomos uma estratégia para particionamento e multiplicação de matrizes por blocos no FPGA virtex 5 (XC5VLX50T -1 FF1136), montado em uma placa (ML 555 Board) da Xilinx

Although most people have experienced word-finding difficulty at one time or another, there are no clinical instruments able to reliably distinguish normal age-related effects from pathology in word-finding impairment. Two experiments were conducted to establish a modified version of the Boston Naming Test (BNT) that includes latency times, the Boston Naming Test of Latencies (BNT-L), in order to improve the instrument's sensitivity to mild to moderate word-finding impairment. Experiment 1: Latency times on the 60-item BNT (Goodglass et al., 2001) for 235 healthy adults' ages 18-89 years were collected on a representative sample. Qualitative features of the BNT items, statistical analyses, IRT, and demographic considerations of age, gender, education, vocabulary, race and culture, helped create a reduced BNT-L version with 15 of the most discriminating items. Statistically sound and sophisticated normative tables are provided that adjust for unseen covariates. Response latencies did not indicate earlier age-related decline in an optimally healthy sample. Experiment 2: Twenty-three patients referred for neuropsychological testing were administered the BNT-L. Patients referred for evaluation of mild cognitive impairment or possible dementia produced significantly different response BNT-L latencies from the healthy sample whereas patients referred for mild brain injury evaluation did not. Normal word-finding problems were discussed in terms of serial stage models of lexical access, as well as in terms of automatic and controlled cognitive processes in younger and older adults. Statistical process for creating a psychometric instrument using latencies is illustrated.

This thesis investigates the effects in speech production of prediction during speech comprehension. The topic is raised by recent theoretical models of speech comprehension, which suggest a more integrated role for speech production and comprehension mechanisms than has previously been posited. The thesis is specifically concerned with the suggestion that during speech comprehension upcoming input is simulated with reference to the listener’s own speech production system by way of efference copy. Throughout this thesis the approach taken is to investigate whether representations elicited during comprehension impact speech production. The representations of interest are those generated endogenously by the listener during prediction of upcoming input. We investigate whether predictions are represented at a form level within the listener’s speech production system. We first present an overview of the relevant literature. We then present details of a picture word interference study undertaken to confirm that the item set employed elicits typical phonological effects within a conventional paradigm in which the competing representation is perceptually available. The main body of the thesis presents evidence concerning the nature of representations arising during prediction, specifically their effect on speech output. We first present evidence from picture naming vocal response latencies. We then complement and extend this with evidence from articulatory imaging, allowing an examination of pre-acoustic aspects of speech production. To investigate effects on speech production as a dynamic motor-activity we employ the Delta method, developed to quantify articulatory variability from EPG and ultrasound recordings. We apply this technique to ultrasound data acquired during mid-sagittal imaging of the tongue and extend the approach to allow us to explore the time-course of articulation during the acoustic response latency period. We investigate whether prediction of another’s speech evokes articulatorily specified activation within the listener’s speech production system The findings presented in this thesis suggest that representations evoked as predictions during speech comprehension do affect speech motor output. However, we found no evidence to suggest that predictions are represented in an articulatorily specified manner. We discuss this conclusion with reference to models of speech production-perception that implicate efference copies in the generation of predictions during speech comprehension.

Designing hardware to output pixels for light field displays or multi-projector systems is challenging owing to the memory bandwidth and speed of the application. A new technique of hardware that implements ‗anywhere pixel routing‘ was designed earlier at the University of Kentucky. This technique uses hardware to route pixels from input to output based upon a Look up Table (LUT). The initial design suffered from high memory latency due to random accesses to the DDR SDRAM input buffer. This thesis presents a cache design that alleviates the memory latency issue by reducing the number of random SDRAM accesses. The cache is implemented in the block RAM of a field programmable gate array (FPGA). A number of simulations are conducted to find an efficient cache. It is found that the cache takes only a few kilobits, about 7% of the block RAM and on an average speeds up the memory accesses by 20-30%.

<p>Localization of speech onsets to determine onset latencies is a complicated problem with as many different methods for finding them as there are different areas which use such measurements. A majority of research performed in cognition uses a standard amplitude threshold voice key for estimating the speech onset latencies but a number of studies have shown that this method is incredibly inaccurate and can bias data or produce contradictory results. A number of alternative methods based on modifications to traditional voice-keys have been proposed to deal with the inconsistency although still show a number of deficiencies. Previous work has suggested that switching from the amplitude domain of a signal to the frequency domain a number of the issues present with voice keys can be overcome and when used in conjunction with a number of highly sensitive heuristics highly accurate onset latencies can be produced reliably under ideal conditions. This research is refined and paired with a new user interface to improve the ease of use and increase the adoption rate of this type of analysis. Recent work in the telecommunications industry also suggests that wavelet-based algorithms in conjunction with the Teager Energy Operator (TEO) can accurately detect speech even in the presence of noise. Four wavelet-based methods are investigated and tested; a simple wavelet transform test, and three methods using wavelet-packet transforms in conjunction with the TEO. Although these methods do not perform very well compared to traditional methods a number of potential issues with the implementation are discussed.</p> / Master of Science (MSc)

The purpose of this thesis is to develop and understand dance balance characteristics on various kinds of dance, related positions and shoe types which contribute to dance performance and to understand different balance controls in various groups. The first study was conducted to examine balance into ballet 2nd position between Ballet shoes and Pointe shoes. Eight dancers performed five different conditions in ballet 2nd position (Ballet Flat, Ballet Demi, Pointe Flat, Pointe Demi, and Pointe Toe) and Centre of Pressure (COP) was used to assess balance. The second study was testing balance control and response to perturbations whilst standing on double leg stance dance positions using a moving platform. Four dance positions were performed by eight subjects (Normal Flat, Turnout Flat, Normal Demi and Turnout Demi) and the platform was moved in two different directions (Forward and Backward) at two different speeds (slow and fast). Kinetics, Kinematics and EMG data was taken from this study. The third study was taken to compare balance control and response to perturbations in single leg standing dance positions between eight dancers and eight gymnasts. The subjects performed static single leg balance in Normal Flat, Turnout Flat, Normal Demi and Turnout Demi. Also, perturbed stance trials were collected in anterior, posterior, right and left directions for two dance positions (Normal Flat and Turnout Flat) at two different speeds (slow and fast) on the moving platform. The results from the studies indicate that dancers move in Medial - Lateral direction more than in Anterior - Posterior direction on Demi-pointe and Toe standing by performing plantar flexion during ballet 2nd position. Demi-pointe position may cause longer delay of EMG latencies because CNS is probably sending information already to keep correcting balance on Demi-pointe. Dancers and Gymnasts have different balance controls due to their ways of training in their performance. Dancers generally reacted faster with slow perturbation in Turnout stance than Gymnasts because this is the particular condition which Dancers are training in.

This thesis addresses decision mechanisms in foraging situations, using laboratory experiments with European starlings (Sturnus vulgaris). Building on previous work from the Behavioural Ecology Research Group, I chose the Sequential Choice Model (SCM; reviewed in Kacelnik et al., 2011 − Appendix 1) as a starting point, and tested its premises and predictions generalising it to different experimental protocols. Classical decision models do not relate choice preferences to behaviour towards isolated options, and assume that choices involve time-consuming evaluations of all alternatives. However, previous work found that starlings’ responses to isolated options predict preference in choices, and that response times to single-option encounters are not reliably longer than response times in choices. Since, in the wild, options are normally encountered sequentially, dealing with isolated options can be considered of greater biological, and possibly psychological, significance than simultaneous decisions. Following this rationale, the SCM postulates that when multiple simultaneous stimuli are met they are processed in parallel, each competing against the memory of background opportunities, rather than comparing present options to each other. At the time of launching this research, these ideas had only been applied to protocols involving just two deterministic alternatives and offering no chance to explore the influence of learning history (i.e., how animals learn to choose; see Chapter 4). To increase their relevance and offer more rigorous tests, I generalised them to situations with multiple (see Chapters 2, 4 and 5), and in some cases probabilistic alternatives (see Chapter 3), controlling the learning regime. I combined these extensions with tests of economic rationality (see Chapter 6), a concept that is presently facing sustained debates. Integrating the result of all experimental chapters (see Chapter 7), my results support the notion that behaviour in single-option encounters is fundamental to understand choice behaviour. The important issue of whether choices involve a decision time cost or the opposite, a shortening of response times, remains unsolved, as neither could be evidenced reliably.

No description available.

Professional audio processing systems such as digital musical instruments, audiomixers, etc. must operate with very tight constraints on overall processing latencyand CPU performance. Consequently, traditional implementations are still mostlybased on specialized hardware like Digital Signal Processors (DSP) and Real-TimeOperating Systems (RTOS) to meet such requirements. However, such systems areminimalistic in nature and they lack many features (e.g. network connectivity, widehardware support, etc.) that a general-purpose operating system such as Linuxoffers. Linux is a very popular choice for the operating system used in embeddeddevices, and many developers have started to use it for designing real-time systemswith relaxed timing constraints. However, none of the available solutions using astandard Linux kernel can satisfy the low-latency requirements of professional audiosystems.In this thesis, a dual kernel approach is employed to enable an embedded Linuxsystem to process audio with low roundtrip latency. The solution is developed usingthe Xenomai framework for real-time computation, which is based on a techniqueknown as interrupt pipeline (I-pipe). I-Pipe enables interrupt virtualization througha micro-kernel running between the Linux kernel and the interrupt controller hardware.The designed system includes an x86 Atom System-on-Chip (SoC), an XMOSmicrocontroller and audio converters to and from the analog domain. Custom kerneldrivers and libraries have been developed to expose the audio programming functionalitiesto programs running in user-space. As a result, the system can achieverobust real-time performance appropriate for professional audio applications, andat the same time it has all the advantages of a traditional Linux solution such ascompatibility with external devices and ease of programming. The real-time capabilityis measured by evaluating the performance in terms of worst case responsetime of the real-time tasks in comparison to the same metrics obtained under astandard Linux kernel. The overall roundtrip latency of audio processing is showedto be improved by almost an order of magnitude (around 2.5ms instead of 20ms). / Profesionella system för ljudbearbetning, som digitala musikinstrument, mixerbord,etc, arbetar med väldigt hårda krav på tidfördröjning och CPU-prestanda. Som enkonsekvens har dessa system traditionellt implementerats på specialiserad hårdvarasom specifika DSP-processor och speciella realtidsoperativsystem. Den typen avsystem är till sin natur minimalistiska och saknar många funktioner (till exempelnätverk och brett stöd för olika hårdvaror) som mer generella operativsystem,som Linux, kan erbjuda. Linux är ett väldigt populärt val av operativsystem förinbyggda system och många utvecklare har även börjat använda det till realtidssystemmed mindre hårda tidskrav. Det finns dock idag inte någon lösning med enstandard-linuxkärna som kan tillfredsställda de krav på låg fördröjning som krävsför användning i profesionella ljudsystem.I det här examensarbetet används en dubbelkärneuppsättning för att ge ettinbyggt Linuxsystem möjlighet att bearbeta digitalt ljud med låg fördröjning. Lösningenanvänder Xenomai-ramverket för realtidsberäkningar baserat på en teknikkallad interrupt pipeline (I-pipe). I-pipe ger möjlighet att virtualisera interruptgenom en mikrokärna som körs som ett lager mellan Linuxkärnan och hårdvaransinterruptcontroller.Det resulterande systemet inkluderar ett x86 Atom-enchipssystem, en XMOSmicrocontroller, och ljudkonverterare till och från analoga ljud in- och utgångar.Drivrutiner och bibliotek utvecklas för att ge direkt tillgång till ljudfunktionerfrån applikationer. Systemet ges därmed robust realtidsprestanda som gör detlämpligt för profesionella ljudtillämpningar samtidigt som det behåller alla fördelarfrån ett traditionellt Linuxsystem, som kompabilitet med extern hårdvara och enklareapplikationsutveckling. Systemets realtidsprestanda utvärderas som den maximalauppmätta tidfördröjning vid realtidsberäkningar jämfört med motsvarandeberäkningar på en standardlinuxkärna. Resultaten visade på en förbättring på nästanen storleksordning (ca 2,5ms mot 20ms).