Relative entropy rate based design for linear hybrid system models

Hybrid system representations have been applied to many challenging modeling situations. In these hybrid system representations, a mixture of continuous and discrete states is used to capture the dominating behavioural features of a nonlinear, possible uncertain, model under approximation. Unfortunately, the problem of how to best design a suitable hybrid system model has not yet been fully addressed. This paper proposes a new joint state measurement relative entropy rate based approach for this design purpose. Design examples and simulation studies are presented which highlight the benefits of our proposed design approaches.

Platinum/graphene nanosheet/SiC contacts and their application for hydrogen gas sensing

Pt/graphene nanosheet/SiC based devices are fabricated and characterized and their performances toward hydrogen gas are investigated. The graphene nanosheets are synthesized via the reduction of spray-coated graphite oxide deposited onto SiC substrates. Raman and X-ray photoelectron spectroscopies indicate incomplete reduction of the graphite oxide, resulting in partially oxidized graphene nanosheet layers of less than 10 nm thickness. The effects of interfaces on the nonlinear behavior of the Pt/graphene and graphene/SiC junctions are investigated. Current-voltage measurements of the sensors toward 1% hydrogen in synthetic air gas mixture at various temperatures ranging up to 100. ° C are performed. From the dynamic response, a voltage shift of ∼100 mV is recorded for 1% hydrogen at a constant current bias of 1 mA at 100. °C. © 2010 American Chemical Society.

Application of higher order spectra to identify epileptic EEG

Epilepsy is characterized by the spontaneous and seemingly unforeseeable occurrence of seizures, during which the perception or behavior of patients is disturbed. An automatic system that detects seizure onsets would allow patients or the people near them to take appropriate precautions, and could provide more insight into this phenomenon. Various methods have been proposed to predict the onset of seizures based on EEG recordings. The use of nonlinear features motivated by the higher order spectra (HOS) has been reported to be a promising approach to differentiate between normal, background (pre-ictal) and epileptic EEG signals. In this work, we made a comparative study of the performance of Gaussian mixture model (GMM) and Support Vector Machine (SVM) classifiers using the features derived from HOS and from the power spectrum. Results show that the selected HOS based features achieve 93.11% classification accuracy compared to 88.78% with features derived from the power spectrum for a GMM classifier. The SVM classifier achieves an improvement from 86.89% with features based on the power spectrum to 92.56% with features based on the bispectrum.

Vector quantization based Gaussian modeling for speaker verification

Gaussian mixture models (GMMs) have become an established means of modeling feature distributions in speaker recognition systems. It is useful for experimentation and practical implementation purposes to develop and test these models in an efficient manner particularly when computational resources are limited. A method of combining vector quantization (VQ) with single multi-dimensional Gaussians is proposed to rapidly generate a robust model approximation to the Gaussian mixture model. A fast method of testing these systems is also proposed and implemented. Results on the NIST 1996 Speaker Recognition Database suggest comparable and in some cases an improved verification performance to the traditional GMM based analysis scheme. In addition, previous research for the task of speaker identification indicated a similar system perfomance between the VQ Gaussian based technique and GMMs

An approach to statistical lip modelling for speaker identification via chromatic feature extraction

This paper presents a novel technique for the tracking of moving lips for the purpose of speaker identification. In our system, a model of the lip contour is formed directly from chromatic information in the lip region. Iterative refinement of contour point estimates is not required. Colour features are extracted from the lips via concatenated profiles taken around the lip contour. Reduction of order in lip features is obtained via principal component analysis (PCA) followed by linear discriminant analysis (LDA). Statistical speaker models are built from the lip features based on the Gaussian mixture model (GMM). Identification experiments performed on the M2VTS¹ database, show encouraging results

New variational Bayesian approaches for statistical data mining : with applications to profiling and differentiating habitual consumption behaviour of customers in the wireless telecommunication industry

This thesis investigates profiling and differentiating customers through the use of statistical data mining techniques. The business application of our work centres on examining individuals’ seldomly studied yet critical consumption behaviour over an extensive time period within the context of the wireless telecommunication industry; consumption behaviour (as oppose to purchasing behaviour) is behaviour that has been performed so frequently that it become habitual and involves minimal intentions or decision making. Key variables investigated are the activity initialised timestamp and cell tower location as well as the activity type and usage quantity (e.g., voice call with duration in seconds); and the research focuses are on customers’ spatial and temporal usage behaviour. The main methodological emphasis is on the development of clustering models based on Gaussian mixture models (GMMs) which are fitted with the use of the recently developed variational Bayesian (VB) method. VB is an efficient deterministic alternative to the popular but computationally demandingMarkov chainMonte Carlo (MCMC) methods. The standard VBGMMalgorithm is extended by allowing component splitting such that it is robust to initial parameter choices and can automatically and efficiently determine the number of components. The new algorithm we propose allows more effective modelling of individuals’ highly heterogeneous and spiky spatial usage behaviour, or more generally human mobility patterns; the term spiky describes data patterns with large areas of low probability mixed with small areas of high probability. Customers are then characterised and segmented based on the fitted GMM which corresponds to how each of them uses the products/services spatially in their daily lives; this is essentially their likely lifestyle and occupational traits. Other significant research contributions include fitting GMMs using VB to circular data i.e., the temporal usage behaviour, and developing clustering algorithms suitable for high dimensional data based on the use of VB-GMM.

Cross likelihood ratio based speaker clustering using eigenvoice models

This paper proposes the use of eigenvoice modeling techniques with the Cross Likelihood Ratio (CLR) as a criterion for speaker clustering within a speaker diarization system. The CLR has previously been shown to be a robust decision criterion for speaker clustering using Gaussian Mixture Models. Recently, eigenvoice modeling techniques have become increasingly popular, due to its ability to adequately represent a speaker based on sparse training data, as well as an improved capture of differences in speaker characteristics. This paper hence proposes that it would be beneficial to capitalize on the advantages of eigenvoice modeling in a CLR framework. Results obtained on the 2002 Rich Transcription (RT-02) Evaluation dataset show an improved clustering performance, resulting in a 35.1% relative improvement in the overall Diarization Error Rate (DER) compared to the baseline system.

Visual voice activity detection using frontal versus profile views

Visual activity detection of lip movements can be used to overcome the poor performance of voice activity detection based solely in the audio domain, particularly in noisy acoustic conditions. However, most of the research conducted in visual voice activity detection (VVAD) has neglected addressing variabilities in the visual domain such as viewpoint variation. In this paper we investigate the effectiveness of the visual information from the speaker’s frontal and profile views (i.e left and right side views) for the task of VVAD. As far as we are aware, our work constitutes the first real attempt to study this problem. We describe our visual front end approach and the Gaussian mixture model (GMM) based VVAD framework, and report the experimental results using the freely available CUAVE database. The experimental results show that VVAD is indeed possible from profile views and we give a quantitative comparison of VVAD based on frontal and profile views The results presented are useful in the development of multi-modal Human Machine Interaction (HMI) using a single camera, where the speaker’s face may not always be frontal.

Preparing the actor for performance capture

Notions of capture and manipulation infer the existence of an interface that combines performer with system, an interface that separates (or intervenes in) the space between performance, intention and history. It is precisely the effect and provided opportunity of the intermediary device on the practice and craft of the actor, that is particularly examined in this work. Defining the scope of current practice for the contemporary actor is a key construct of this challenge with the most appropriate definition revolving around the pursuit of providing a required mixture of performance and content for live, mediated, framed and variously captured formats that exist in the present day. One of these particular formats is Performance Capture which this paper interrogate in more detail.

Evaluating the transport impacts of transit oriented developments (TODS)

Sustainable transport has become a necessity instead of an option, to address the problems of congestion and urban sprawl, whose effects include increased trip lengths and travel time. A more sustainable form of development, known as Transit Oriented Development (TOD) is presumed to offer sustainable travel choices with reduced need to travel to access daily destinations, by providing a mixture of land uses together with good quality of public transport service, infrastructure for walking and cycling. However, performance assessment of these developments with respect to travel characteristics of their inhabitants is required. This research proposes a five step methodology for evaluating the transport impacts of TODs. The steps for TOD evaluation include pre–TOD assessment, traffic and travel data collection, determination of traffic impacts, determination of travel impacts, and drawing outcomes. Typically, TODs are comprised of various land uses; hence have various types of users. Assessment of characteristics of all user groups is essential for obtaining an accurate picture of transport impacts. A case study TOD, Kelvin Grove Urban Village (KGUV), located 2km of north west of the Brisbane central business district in Australia was selected for implementing the proposed methodology and to evaluate the transport impacts of a TOD from an Australian perspective. The outcomes of this analysis indicated that KGUV generated 27 to 48 percent less traffic compared to standard published rates specified for homogeneous uses. Further, all user groups of KGUV used more sustainable modes of transport compared to regional and similarly located suburban users, with higher trip length for shopping and education trips. Although the results from this case study development support the transport claims of reduced traffic generation and sustainable travel choices by way of TODs, further investigation is required, considering different styles, scales and locations of TODs. The proposed methodology may be further refined by using results from new TODs and a framework for TOD evaluation may be developed.

Centrifuging rotary vacuum filter mud at Tully Mill

Disposal of mud and ash, particularly in wet weather conditions, is a significant expense for mills. This paper reports on one part of a process to pelletise mud and ash, aimed at making mud and ash more attractive to growers across entire mill districts. The full process is described in a separate paper. The part described in this paper involves re-constituting mud cake from the filter station at Tully Mill and processing it in a decanter centrifuge. The material produced by re-constituting and centrifuging is drier and made up of separate particles. The material needs to mix easily with boiler ash, and the mixture needs to be fed easily into a flue gas drier to be dried to low moisture. The results achieved with the particular characteristics of Tully Mill rotary vacuum filter cake are presented. It was found that an internal rotor with a 20º beach was not adequate to process re-constituted rotary vacuum filter mud. A rotor with a 10º beach worked much more successfully. A total of four tonnes of centrifuged mud with a moisture content ranging from 60% to 65% was produced. It was found that the torque, flocculant rate and dose rate had a statistically significant effect on the moisture content. Feed rate did not have a noticeable impact on the moisture content by itself but torque had a much larger impact on the moisture content at the low feed rate than at the high feed rate. These results indicated that the moisture content of the mud can most likely be reduced with low feed rate, low flocculant rate, high dose rate and high torque. One issue that is believed to affect the operation of a decanter centrifuge was the large quantity of long bagasse fibres in the rotary vacuum filter mud. It is likely that the long fibres limited the throughput of the centrifuge and the moisture achieved.