Travel time prediction on signalised urban arterials by applying SARIMA modelling on Bluetooth data

Travel time prediction has long been the topic of transportation research. But most relevant prediction models in the literature are limited to motorways. Travel time prediction on arterial networks is challenging due to involving traffic signals and significant variability of individual vehicle travel time. The limited availability of traffic data from arterial networks makes travel time prediction even more challenging. Recently, there has been significant interest of exploiting Bluetooth data for travel time estimation. This research analysed the real travel time data collected by the Brisbane City Council using the Bluetooth technology on arterials. Databases, including experienced average daily travel time are created and classified for approximately 8 months. Thereafter, based on data characteristics, Seasonal Auto Regressive Integrated Moving Average (SARIMA) modelling is applied on the database for short-term travel time prediction. The SARMIA model not only takes the previous continuous lags into account, but also uses the values from the same time of previous days for travel time prediction. This is carried out by defining a seasonality coefficient which improves the accuracy of travel time prediction in linear models. The accuracy, robustness and transferability of the model are evaluated through comparing the real and predicted values on three sites within Brisbane network. The results contain the detailed validation for different prediction horizons (5 min to 90 minutes). The model performance is evaluated mainly on congested periods and compared to the naive technique of considering the historical average.

Sequential decision fusion of multibiometrics applied to text-dependent speaker verification for controlled errors

Reliability of the performance of biometric identity verification systems remains a significant challenge. Individual biometric samples of the same person (identity class) are not identical at each presentation and performance degradation arises from intra-class variability and inter-class similarity. These limitations lead to false accepts and false rejects that are dependent. It is therefore difficult to reduce the rate of one type of error without increasing the other. The focus of this dissertation is to investigate a method based on classifier fusion techniques to better control the trade-off between the verification errors using text-dependent speaker verification as the test platform. A sequential classifier fusion architecture that integrates multi-instance and multisample fusion schemes is proposed. This fusion method enables a controlled trade-off between false alarms and false rejects. For statistically independent classifier decisions, analytical expressions for each type of verification error are derived using base classifier performances. As this assumption may not be always valid, these expressions are modified to incorporate the correlation between statistically dependent decisions from clients and impostors. The architecture is empirically evaluated by applying the proposed architecture for text dependent speaker verification using the Hidden Markov Model based digit dependent speaker models in each stage with multiple attempts for each digit utterance. The trade-off between the verification errors is controlled using the parameters, number of decision stages (instances) and the number of attempts at each decision stage (samples), fine-tuned on evaluation/tune set. The statistical validation of the derived expressions for error estimates is evaluated on test data. The performance of the sequential method is further demonstrated to depend on the order of the combination of digits (instances) and the nature of repetitive attempts (samples). The false rejection and false acceptance rates for proposed fusion are estimated using the base classifier performances, the variance in correlation between classifier decisions and the sequence of classifiers with favourable dependence selected using the 'Sequential Error Ratio' criteria. The error rates are better estimated by incorporating user-dependent (such as speaker-dependent thresholds and speaker-specific digit combinations) and class-dependent (such as clientimpostor dependent favourable combinations and class-error based threshold estimation) information. The proposed architecture is desirable in most of the speaker verification applications such as remote authentication, telephone and internet shopping applications. The tuning of parameters - the number of instances and samples - serve both the security and user convenience requirements of speaker-specific verification. The architecture investigated here is applicable to verification using other biometric modalities such as handwriting, fingerprints and key strokes.

A Bayesian regression approach to assess uncertainty in pollutant wash-off modelling

Due to knowledge gaps in relation to urban stormwater quality processes, an in-depth understanding of model uncertainty can enhance decision making. Uncertainty in stormwater quality models can originate from a range of sources such as the complexity of urban rainfall-runoff-stormwater pollutant processes and the paucity of observed data. Unfortunately, studies relating to epistemic uncertainty, which arises from the simplification of reality are limited and often deemed mostly unquantifiable. This paper presents a statistical modelling framework for ascertaining epistemic uncertainty associated with pollutant wash-off under a regression modelling paradigm using Ordinary Least Squares Regression (OLSR) and Weighted Least Squares Regression (WLSR) methods with a Bayesian/Gibbs sampling statistical approach. The study results confirmed that WLSR assuming probability distributed data provides more realistic uncertainty estimates of the observed and predicted wash-off values compared to OLSR modelling. It was also noted that the Bayesian/Gibbs sampling approach is superior compared to the most commonly adopted classical statistical and deterministic approaches commonly used in water quality modelling. The study outcomes confirmed that the predication error associated with wash-off replication is relatively higher due to limited data availability. The uncertainty analysis also highlighted the variability of the wash-off modelling coefficient k as a function of complex physical processes, which is primarily influenced by surface characteristics and rainfall intensity.

Variability of metal composition and concentrations in road dust in the urban environment

Urban road dust comprises of a range of potentially toxic metal elements and plays a critical role in degrading urban receiving water quality. Hence, assessing the metal composition and concentration in urban road dust is a high priority. This study investigated the variability of metal composition and concentrations in road dust in 4 different urban land uses in Gold Coast, Australia. Samples from 16 road sites were collected and tested for selected 12 metal species. The data set was analyzed using both univariate and multivariate techniques. Outcomes of the data analysis revealed that the metal concentrations in road dust differ considerably within and between different land uses. Iron, aluminum, magnesium and zinc are the most abundant in urban land uses. It was also noted that metal species such as titanium, nickel, copper and zinc have the highest concentrations in industrial land use. The study outcomes revealed that soil and traffic related sources as key sources of metals deposited on road surfaces.

Modelling after-the-fact leakage for key exchange

Security models for two-party authenticated key exchange (AKE) protocols have developed over time to prove the security of AKE protocols even when the adversary learns certain secret values. In this work, we address more granular leakage: partial leakage of long-term secrets of protocol principals, even after the session key is established. We introduce a generic key exchange security model, which can be instantiated allowing bounded or continuous leakage, even when the adversary learns certain ephemeral secrets or session keys. Our model is the strongest known partial-leakage-based security model for key exchange protocols. We propose a generic construction of a two-pass leakage-resilient key exchange protocol that is secure in the proposed model, by introducing a new concept: the leakage-resilient NAXOS trick. We identify a special property for public-key cryptosystems: pair generation indistinguishability, and show how to obtain the leakage-resilient NAXOS trick from a pair generation indistinguishable leakage-resilient public-key cryptosystem.

Characterising an ECG signal using statistical modelling : a feasibility study

For clinical use, in electrocardiogram (ECG) signal analysis it is important to detect not only the centre of the P wave, the QRS complex and the T wave, but also the time intervals, such as the ST segment. Much research focused entirely on qrs complex detection, via methods such as wavelet transforms, spline fitting and neural networks. However, drawbacks include the false classification of a severe noise spike as a QRS complex, possibly requiring manual editing, or the omission of information contained in other regions of the ECG signal. While some attempts were made to develop algorithms to detect additional signal characteristics, such as P and T waves, the reported success rates are subject to change from person-to-person and beat-to-beat. To address this variability we propose the use of Markov-chain Monte Carlo statistical modelling to extract the key features of an ECG signal and we report on a feasibility study to investigate the utility of the approach. The modelling approach is examined with reference to a realistic computer generated ECG signal, where details such as wave morphology and noise levels are variable.

Computation of ECG signal features using MCMC modelling in software and FPGA reconfigurable hardware

Computational optimisation of clinically important electrocardiogram signal features, within a single heart beat, using a Markov-chain Monte Carlo (MCMC) method is undertaken. A detailed, efficient data-driven software implementation of an MCMC algorithm has been shown. Initially software parallelisation is explored and has been shown that despite the large amount of model parameter inter-dependency that parallelisation is possible. Also, an initial reconfigurable hardware approach is explored for future applicability to real-time computation on a portable ECG device, under continuous extended use.

Aligning off-balance sheet risk, on-balance sheet risk and audit fees: a PLS path modelling analysis

This study focuses on using the partial least squares (PLS) path modelling technique in archival auditing research by replicating the data and research questions from prior bank audit fee studies. PLS path modelling allows for inter-correlations among audit fee determinants by establishing latent constructs and multiple relationship paths in one simultaneous PLS path model. Endogeneity concerns about auditor choice can also be addressed with PLS path modelling. With a sample of US bank holding companies for the period 2003-2009, we examine the associations among on-balance sheet financial risks, off-balance sheet risks and audit fees, and also address the pervasive client size effect, and the effect of the self-selection of auditors. The results endorse the dominating effect of size on audit fees, both directly and indirectly via its impacts on other audit fee determinants. By simultaneously considering the self-selection of auditors, we still find audit fee premiums on Big N auditors, which is the second important factor on audit fee determination. On-balance-sheet financial risk measures in terms of capital adequacy, loan composition, earnings and asset quality performance have positive impacts on audit fees. After allowing for the positive influence of on-balance sheet financial risks and entity size on off-balance sheet risk, the off-balance sheet risk measure, SECRISK, is still positively associated with bank audit fees, both before and after the onset of the financial crisis. The consistent results from this study compared with prior literature provide supporting evidence and enhance confidence on the application of this new research technique in archival accounting studies.

Aligning off-balance sheet risk, on-balance sheet risk and audit fees: a PLS path modelling analysis

This study focuses on using the partial least squares (PLS) path modelling methodology in archival auditing research by replicating the data and research questions from prior bank audit fee studies. PLS path modelling allows for inter-correlations among audit fee determinants by establishing latent constructs and multiple relationship paths in one simultaneous PLS path model. Endogeneity concerns about auditor choice can also be addressed with PLS path modelling. With a sample of US bank holding companies for the period 2003-2009, we examine the associations among on-balance sheet financial risks, off-balance sheet risks and audit fees, and also address the pervasive client size effect, and the effect of the self-selection of auditors. The results endorse the dominating effect of size on audit fees, both directly and indirectly via its impacts on other audit fee determinants. By simultaneously considering the self-selection of auditors, we still find audit fee premiums on Big N auditors, which is the second important factor on audit fee determination. On-balance-sheet financial risk measures in terms of capital adequacy, loan composition, earnings and asset quality performance have positive impacts on audit fees. After allowing for the positive influence of on-balance sheet financial risks and entity size on off-balance sheet risk, the off-balance sheet risk measure, SECRISK, is still positively associated with bank audit fees, both before and after the onset of the financial crisis. The consistent results from this study compared with prior literature provide supporting evidence and enhance confidence on the application of this new research technique in archival accounting studies.

Short utterance PLDA speaker verification using SN-WLDA and variance modelling techniques

This paper proposes a combination of source-normalized weighted linear discriminant analysis (SN-WLDA) and short utterance variance (SUV) PLDA modelling to improve the short utterance PLDA speaker verification. As short-length utterance i-vectors vary with the speaker, session variations and phonetic content of the utterance (utterance variation), a combined approach of SN-WLDA projection and SUV PLDA modelling is used to compensate the session and utterance variations. Experimental studies have found that a combination of SN-WLDA and SUV PLDA modelling approach shows an improvement over baseline system (WCCN[LDA]-projected Gaussian PLDA (GPLDA)) as this approach effectively compensates the session and utterance variations.

Morphometric modelling of ageing in the human pubic symphysis : sexual dimorphism in an Australian population

Despite the prominent use of the pubic symphysis for age estimation in forensic anthropology, little has been documented regarding the quantitative morphological and micro-architectural changes of this surface. Specifically, utilising post-mortem computed tomography data from a large, contemporary Australian adult population, this study aimed to evaluate sexual dimorphism in the morphology and bone composition of the symphyseal surface; and temporal characterisation of the pubic symphysis in individuals of advancing age. The sample consisted of multi-slice computed tomography (MSCT) scans of the pubic symphysis(slice thickness: 0.5 mm, overlap: 0.1 mm) of 200 individuals of Caucasian ancestry aged 15–70 years, obtained in 2011. Surface rendering reconstruction of the symphyseal surface was conducted in OsiriX1 (v.4.1) and quantitative analyses in Rapidform XOSTM and OsteomeasureTM. Morphometric variables including inter-pubic distance, surface area, circumference, maximum height and width of the symphyseal surface and micro-architectural assessment of cortical and trabecular bone compositions were quantified using novel automated engineering software capabilities. The major results of this study are correlated with the macroscopic ossification and degeneration pattern of the symphyseal surface, demonstrating significant age-related changes in the morphometric and bone tissue variables between 15 and 70 years. Regardless of sex, the overall dimensions of the symphyseal surface increased with age, coupled with a decrease in bone mass in the trabecular and cortical bone compartments. Significant differences between the ventral, dorsal and medial cortical surfaces were observed, which may be correlated to bone formation activity dependent on muscle activity and ligamentous attachments. Our study demonstrates significant sexual dimorphism at this site, with males exhibiting greater surface dimensions than females. These baseline results provide a detailed insight into the changes in the structure of the pubic symphysis with ageing and sexually dimorphic features associated with the cortical and trabecular bone profiles.

Investigations on road noise level spatial variability within a specially designed acoustic balcony

An investigation into the spatial distribution of road traffic noise levels on a balcony is conducted. A balcony constructed to a special acoustic design due to its elevation above an 8 lane motorway is selected for detailed measurements. The as-constructed balcony design includes solid parapets, side walls, ceiling shields and highly absorptive material placed on the ceiling. Road traffic noise measurements are conducted spatially using a five channel acoustic analyzer, where four microphones are located at various positions within the balcony space and one microphone placed outside the parapet at a reference position. Spatial distributions in both vertical and horizontal planes are measured. A theoretical model and prediction configuration is presented that assesses the acoustic performance of the balcony under existing traffic flow conditions. The prediction model implements a combined direct path, specular reflection path and diffuse reflection path utilizing image source and radiosity techniques. Results obtained from the prediction model are presented and compared to the measurement results. The predictions are found to correlate well with measurements with some minor differences that are explained. It is determined that the prediction methodology is acceptable to assess a wider range of street and balcony configuration scenarios.

Improving PLDA speaker verification using WMFD and linear-weighted approaches in limited microphone data conditions

This paper proposes the addition of a weighted median Fisher discriminator (WMFD) projection prior to length-normalised Gaussian probabilistic linear discriminant analysis (GPLDA) modelling in order to compensate the additional session variation. In limited microphone data conditions, a linear-weighted approach is introduced to increase the influence of microphone speech dataset. The linear-weighted WMFD-projected GPLDA system shows improvements in EER and DCF values over the pooled LDA- and WMFD-projected GPLDA systems in inter-view-interview condition as WMFD projection extracts more speaker discriminant information with limited number of sessions/ speaker data, and linear-weighted GPLDA approach estimates reliable model parameters with limited microphone data.

Seasonal and spatial variations in rare earth elements to identify inter-aquifer linkages and recharge processes in an Australian catchment

With the aim of elucidating the seasonal behaviour of rare earth elements (REEs), surface and groundwaters were collected under dry and wet conditions in different hydrological units of the Teviot Brook catchment (Southeast Queensland, Australia). Sampled waters showed a large degree of variability in both REE abundance and normalised patterns. Overall REE abundance ranged over nearly three orders of magnitude, and was consistently lower in the sedimentary bedrock aquifer (18ppt<∑REE<477ppt) than in the other hydrological systems studied. Abundance was greater in springs draining rhyolitic rocks (∑REE=300 and 2054ppt) than in springs draining basalt ranges (∑REE=25 and 83ppt), yet was highly variable in the shallow alluvial groundwater (16ppt<∑REE<5294ppt) and, to a lesser extent, in streamwater (85ppt<∑REE<2198ppt). Generally, waters that interacted with different rock types had different REE patterns. In order to obtain an unbiased characterisation of REE patterns, the ratios between light and middle REEs (R(M/L)) and the ratios between middle and heavy REEs (R(H/M)) were calculated for each sample. The sedimentary bedrock aquifer waters had highly evolved patterns depleted in light REEs and enriched in middle and heavy REEs (0.17intercepting the sedimentary bedrock, where the freshly recharged waters likely contributed to the deeper groundwater flow during the wet season (decrease from R(M/L)=0.81 and 0.56 to R(M/L)=0.46 and 0.17). Results from this study suggest that REEs may be usefully applied as indicators of recharge processes and inter-aquifer mixing. They also underline the importance of conducting seasonal sampling campaigns to capture possible short-term variations in REE patterns and abundance, which is essential to enable a better understanding of hydrological and hydrochemical processes in complex geological settings