The thesis illustrates a DSP-based speaker recognition system . In order to make the modular within the representation floating-point, we simplify the algorithm. This speaker recognition system is including hardware setting and implementation of speaker algorithm. The DSP chip is float arithmetic DSP(ADSP-21161 of ADI SHARK Series) , the algorithm of speaker recognition is gaussian mixture model. According to result of experiments, the speaker recognition of DSP can gain good recognition and speed efficiency.

Recent research suggests that amino acids, such as leucine and isoleucine, can improve glucose tolerance in vivo and in vitro animal models by accelerating glucose uptake in peripheral tissues and stimulate glycogen synthesis in vitro in the absence of insulin. Our laboratory recently found that gavaging normal Sprague-Dawley rats with an amino acid mixture, composed of isoleucine, leucine, cystine, methionine, and valine, improved blood glucose response during an oral glucose challenge without an increase in the plasma insulin response. The blood glucose-lowering effect of the amino acid mixture was due to an increase in skeletal muscle glucose uptake. These results suggest that this amino acid supplement acutely improves muscle insulin sensitivity and blood glucose homeostasis. However, the effect of this amino acid mixture on glucose tolerance and muscle glycogen synthesis in humans has not been investigated. Some studies have also shown that daily supplementation or acute ingestion of amino acids may prevent muscle damage that occurs as a result of a prolonged, intense endurance exercise or strength training and therefore improves force production and exercise performance. However, the effects of the addition of an amino acid mixture to carbohydrate supplement on muscle damage after a prolonged endurance exercise, as well as on the subsequent anaerobic exercise performance, have not been characterized. Therefore, in this series of two studies, the effects of an amino acid mixture, composed of isoleucine, leucine, cyctine, methionine, and valine, on glucose tolerance, muscle glycogen resynthesis, muscle damage, and anaerobic exercise performance were investigated. Study 1 demonstrated that our amino acid mixture lowered the glucose response to an OGTT in healthy overweight/obese subjects in an insulin-independent manner. Study 2 demonstrated that both high and low dosages of amino acid mixture were effective in lowering blood glucose response to a carbohydrate bolus in athletes postexercise. High dosage of amino acid mixture was more potent in glucose regulation by providing a higher insulin response and amino acid effect. However, our amino acid mixture had no effects on post exercise muscle glycogen synthesis, exercise-induced muscle damage or subsequent anaerobic performance. Taken together, the results of this research series suggest that an amino acid mixture, composed of isoleucine and 4 additional amino acids, attenuates the glucose response to a glucose bolus in an insulin-independent manner, but does not enhance muscle glycogen restoration following exercise or prevent exercise-induced muscle damage. / text

Differential Item Functioning (DIF) occurs when examinees with the same ability have different probabilities of endorsing an item. Conventional DIF detection methods (e.g., the Mantel-Hansel test) can be used to detect DIF only across observed groups, such as gender or ethnicity. However, research has found that DIF is not typically fully explained by an observed variable (e.g., Cohen & Bolt, 2005). True source of DIF may be unobserved, including variables such as personality, response patterns, or unmeasured background variables. The Factor Mixture Model (FMM) is designed to detect unobserved sources of heterogeneity in factor structures, and an FMM with binary outcomes has recently been used for assessing DIF (DeMars & Lau, 2011; Jackman, 2010). However, FMMs with binary outcomes for detecting DIF have not been thoroughly explored to investigate both types of between-class latent DIF (LDIF) and class-specific observed DIF (ODIF). The present simulation study was designed to investigate whether models correctly specified in terms of LDIF and/or ODIF influence the performance of model fit indices (AIC, BIC, aBIC, and CAIC) and entropy, as compared to models incorrectly specified in terms of either LDIF or ODIF. In addition, the present study examined the recovery of item difficulty parameters and investigated the proportion of replications in which items were correctly or incorrectly identified as displaying DIF, by manipulating DIF effect size and latent class probability. For each simulation condition, two latent classes of 27 item responses were generated to fit a one parameter logistic model with items’ difficulties generated to exhibit DIF across the classes and/or the observed groups. Results showed that FMMs with binary outcomes performed well in terms of fit indices, entropy, DIF detection, and recovery of large DIF effects. When class probabilities were unequal with small DIF effects, performance decreased for fit indices, power, and the recovery of DIF effects compared to equal class probability conditions. Inflated Type I errors were found for invariant DIF items across simulation conditions. When data were generated to fit a model having ODIF but estimated LDIF, specifying LDIF in the model fully captured ODIF effects when DIF effect sizes were large. / text

Single nucleotide polymorphisms (SNPs) have been increasingly popular for a wide range of genetic studies. A high-throughput genotyping technologies usually involves a statistical genotype calling algorithm. Most calling algorithms in the literature, using methods such as k-means and mixturemodels, rely on elliptical structures of the genotyping data; they may fail when the minor allele homozygous cluster is small or absent, or when the data have extreme tails or linear patterns. We propose an automatic genotype calling algorithm by further developing a linear grouping algorithm (Van Aelst et al., 2006). The proposed algorithm clusters unnormalized data points around lines as against around centroids. In addition, we associate a quality value, silhouette width, with each DNA sample and a whole plate as well. This algorithm shows promise for genotyping data generated from TaqMan technology (Applied Biosystems). A key feature of the proposed algorithm is that it applies to unnormalized fluorescent signals when the TaqMan SNP assay is used. The algorithm could also be potentially adapted to other fluorescence-based SNP genotyping technologies such as Invader Assay. Motivated by the SNP genotyping problem, we propose a partial likelihood approach to linear clustering which explores potential linear clusters in a data set. Instead of fully modelling the data, we assume only the signed orthogonal distance from each data point to a hyperplane is normally distributed. Its relationships with several existing clustering methods are discussed. Some existing methods to determine the number of components in a data set are adapted to this linear clustering setting. Several simulated and real data sets are analyzed for comparison and illustration purpose. We also investigate some asymptotic properties of the partial likelihood approach. A Bayesian version of this methodology is helpful if some clusters are sparse but there is strong prior information about their approximate locations or properties. We propose a Bayesian hierarchical approach which is particularly appropriate for identifying sparse linear clusters. We show that the sparse cluster in SNP genotyping datasets can be successfully identified after a careful specification of the prior distributions.

This thesis devises quantization and source-channel coding schemes to increase the error robustness of the newly standardized ITU-T G.711.1 speech coder. The schemes employ Gaussian mixture model (GMM) based multiple description quantizers (MDQ). The thesis reviews the literature focusing on GMM based quantization, MDQ, and GMM-MDQ design methods and bit allocation schemes. GMM-MDQ are then designed for the quantization and coding of the MDCT coefficients in the G.711.1 speech coder. The designs are optimized for and tested over packet erasure channels. Performance of the designs are compared with Mohr's forward error correcting code based multiple description coding (MDC) scheme. / Thesis (Master, Electrical & Computer Engineering) -- Queen's University, 2010-06-02 16:02:11.727

No description available.

The development of radiata pine (Pinus radiata D. Don.) clones was compared in monoclonal and clonal mixture plots planted in an experiment established at Dalethorpe, Canterbury, New Zealand with ten radiata pine clones in September 1993. Clones were deployed in a randomised complete block plot design with three replications. Each replication contained ten treatments of monoclonal plots and one in which all the clones were intimately mixed in equal proportions. Clones significantly differed in initial morphologies, survival and stem slenderness. Sturdiness and initial heights were found to be the best predictors of initial survivals. The study revealed that mode of deployment did not affect overall productivity, but individual clones exhibited significantly different productivities between modes of deployment. All clones contributed similarly to overall productivity in the monoclonal mode of deployment, whereas the contribution of clones in the clonal mixture mode of deployment was disproportionate. A minority of the clones contributed a majority of overall productivity in the clonal mixture mode of deployment. The inclusion of competition index as an independent variable in a distance-dependent individual tree diameter increment model explained a significant amount of variability in diameter growth. The use of an inverse-squared distance to neighbouring plants in the competition index provided a slightly superior fit to the data compared to one that employed a simple inverse of distance. Addition of genotype information in the competition index further improved the fit of the model. Clones experienced different levels of competition in monoclonal and clonal mixture modes of deployment. Competition in monoclonal plots remained uniform over time, whereas some clones experienced greater competition in clonal mixture plots which led to greater variability in their tree sizes. This study indicated that single tree plot progeny test selections and early selections may miss out some good genotypes that can grow rapidly if deployed monoclonally. Stand level modelling revealed that clones differed significantly in modeled yield patterns and model asymptotes. Clones formed two distinct groups having significantly different yield models. The study also demonstrated that models developed from an initial few years’ data were biased indicators of their relative future performances. Evaluation of effectiveness of the 3-PG hybrid model using parameter values obtained from destructive sampling and species-specific values from different studies revealed that it is possible to calibrate this model for simulating the productivity of clones, and predictions from this model might inform clonal selections at different sites under differing climatic conditions. Destructive sampling at age 5 years revealed that clones significantly differed in foliage and stem biomass. The differences in productivities of clones were mainly due to differences in biomass partitioning and specific leaf areas. Clones significantly differed in dynamic wood stiffness, stem-slenderness, branch diameter, branch index and branch angle at an initial stocking of 1250 stems/ha. Mode of deployment affected stem slenderness, which is sometimes related to stiffness. Although dynamic stiffness was correlated with stem slenderness and stem slenderness exhibited a significant influence on stiffness, clones did not exhibit statistically significant differences in dynamic stiffness. Increasing initial stocking from 833 stems/ha to 2500 stems/ha resulted in a 56 % decrease in branch diameter and a 17 % increase in branch angle. Trees in the monoclonal mode of deployment exhibited greater uniformity with respect to tree size, stem-slenderness, and competition experienced by clones compared to those in the clonal mixture mode of deployment. Susceptibility of one clone to Woolly aphid suggested that greater risks were associated with large scale deployment of susceptible clones in a monoclonal mode of deployment. This study also indicated that if the plants were to be deployed in a monoclonal mode then block plot selections would have greater potential to enhance productivity.

Zeotropic mixtures exhibit a temperature glide between the dew and bubble points during condensation. This glide has the potential to increase system efficiency when matched to the thermal sink in power generation, chemical processing, and heating and cooling systems. To understand the coupled heat and mass transfer mechanisms during phase change of high-glide zeotropic mixtures, a comprehensive investigation of the condensation of ammonia and ammonia/water mixtures in small diameter channels was performed. Condensation heat transfer and pressure drop experiments were conducted with ammonia and ammonia/water mixtures. Experiments on ammonia were conducted for varying tube diameters (0.98 < D < 2.16 mm), mass fluxes (75 < G < 225 kg m⁻² s⁻¹) and saturation conditions (30 < Tsat < 60°C). Zeotropic ammonia/water experiments were conducted for multiple tube diameters (0.98 < D < 2.16 mm), mass fluxes (50 < G < 200 kgm⁻² s⁻¹) and bulk ammonia mass fraction (xbulk = 0.8, 0.9, and > 0.96). An experimental methodology and data analysis procedure for evaluating the local condensation heat duty (for incremental ∆q), condensation transfer coefficient (for pure ammonia), apparent heat transfer coefficient (for zeotropic ammonia/water mixtures), and frictional pressure gradient with low uncertainties was developed. A new heat transfer model for condensation of ammonia in mini/microchannels was developed. Using the insights derived from the pure ammonia work, an improved zeotropic condenser design method for high-temperature-glide mixtures in small diameter channels, based on the non-equilibrium film theory, was introduced. The key features of the improved model were the consideration of annular and non-annular flow effects on liquid film transport, including condensate and vapor sensible cooling contributions, and accounting for mini/microchannel effects through the new liquid film correlation. By understanding the behavior of these mixtures in microchannel geometries, highly efficient, compact thermal conversion devices can be developed.

The following study analyzes birthweight-specific neonatal health using a combination of a mixture model and logistic regression: the extended Parametric Mixture of Logistic Regression. The data are collected from the Obstetric database at Muhimbili National Hospital in Dar es Salaam, Tanzania and the years 2009 -2013 are used in the analysis. Due to rounding in the birthweight data a novel method to adjust for rounding when estimating a mixture model is applied. The influence of rounding on the estimates is then investigated. A three-component model is selected. The variables used in the analysis of neonatal health are early neonatal mortality, if the mother has HIV, anaemia, is a private patient and if the neonate is born after 36 completed weeks of gestation. It can be concluded that the mortality rates are high especially for low birthweights (2000 or less) in the estimated first and second components. However, due to wide confidence bounds it is hard to draw conclusions from the data.

This study examined heterogeneity in adolescents experimentation with partnered sexual behaviours. Participants were 88 high school students in Edmonton, Alberta (M age = 16.59, SD = .95). Students completed surveys online once per two months from December, 2008 through December, 2009. Surveys tracked students reports of seven sexual behaviours ranging in intimacy from holding hands to intercourse. Growth mixture models were used to sort students trajectories of sexual behaviours across months into latent classes based on similar profiles. The best-fitting model revealed three distinct classes, labeled inexperienced, experimenting, and experienced. Students classified as inexperienced primarily reported only lower-intimacy, non-genital sexual behaviours across months, and many reported no sexual behaviours. Students classified as experimenting and experienced reported similar levels of higher-intimacy sexual behaviours across months. Most experimenting students behaviours appeared to increase gradually from less to more intimate, whereas experienced students appeared to make abrupt transitions between lower- and higher-intimacy behaviours, month-to-month. Demographic, personal, peer, and family variables provided additional information that increased distinction among classes, and explained residual within-class heterogeneity. The probability of being classified as inexperienced was highest for students who were younger, reported fewer sexually experienced friends, and lower parent behavioural control. Students who reported higher parent behavioural control had the highest probability of being classified as experimenting. Relations between trajectories of sexual behaviour intimacy and risk factors (e.g., later pubertal timing, fewer problem behaviours) and protective-enhancing resources (e.g., higher psychosocial maturity, more intimate friendships) varied across classes. This study shows that there are multiple pathways of experimentation with sexual behaviour in adolescence. Results are consistent both with studies that emphasize the potential for sex in adolescence to be high-risk, and with studies and arguments that emphasize the potential for sex in adolescence to play an important preparatory role toward healthy adult sexual functioning. Theoretical arguments and discussion are guided by a person-centered, developmental systems approach.