Aravind, Alex

This thesis reviews multiple forms of terahertz spectroscopy and imaging, and exploits a combination of aspects from the techniques involved to undertake a novel method. With the use of this method, subsurface interfaces within a practical, insulated system are spatially imaged and identified. Initially, transmission spectroscopy is used to select the insulating material of the system based on favourable dielectric properties, followed by the use of dove prism reflection spectroscopy to understand and qualitatively contrast the responses of the interfaces to reflecting terahertz pulses. After constructing a reflection imaging system, multiple interface materials concealed by insulation are successfully located and identified by coloured overlays in a 100 by 100 image. Further exploration of applications involving the proposed method is also discussed.

Due to the popularity of the JavaScript Object Notation (JSON), a need has arisen for the creation of schema documents for the purpose of validating the content of other JSON documents. Existing automatic schema generation tools, however, have not adequately considered the scenario of an array of JSON objects with different types of structures. These tools work off the assumption that all objects have the same structure, and thus, only generate a single schema combining them together. To address this problem, this thesis looks to improve upon schema generation for heterogeneous JSON arrays. We develop an algorithm to determine a set of keys that identifies what type of structure each element has. These keys are then used as the basis for a schema decision tree. The objective of this tree is to help in the validation process by allowing each element to be compared against a single, more tailored, schema.

The primary objective of this study was to investigate the effects of a 4-week respiratory muscle endurance training (eRMT) program on the physiological and psychological aspects of central fatigue using, respectively, near-infrared spectroscopy (NIRS) and quantification of effort perceptions during maximal exercise. A secondary objective was to assess any impact of eRMT on respiratory health and exercise performance. This study compared pre- and post-eRMT data from the same group of healthy adults. The results indicated that eRMT did not have any effect on respiratory function, exercise time to exhaustion, or physiological responses to exercise but significantly decreased ratings of perceived exertion (RPE) during exercise. An increase in the concentrations of oxygenated hemoglobin [O2Hb], deoxygenated hemoglobin [HHb], and total hemoglobin [tHb] during exercise was observed post-eRMT compared to pre-eRMT, and this increase differed by hemisphere. Based on these preliminary findings, we suggest an eRMTinduced left-to-right hemodynamic shift during exercise, consistent with the change from a novel to a learned task.

Transcutaneous near infrared spectroscopy (NIRS) of muscle requires coupling between the device and the skin. An unfortunate by-product of this coupling is contact force artefact, where the amount of contact force between the device and the skin affects measurements. Contact force artefact is well known, but largely ignored in most NIRS research. We performed preliminary investigations of contact force artefact to quantify tissue behaviour to inform future NIRS designs. Specifically, we conducted three studies on contact force artefact: (i) an experimental investigation of static load at varied levels of contact force and muscle activation, (ii) an experimental investigation of oscillating load at varied levels of contact force and frequency, and (iii) a Monte Carlo simulation of photon propagation through skin, adipose tissue, and muscle. Our results confirmed that contact force artefact is a confounding factor in NIRS muscle measurements because contact force affects measured hemoglobin concentrations in a manner consistent with muscle contractions. Further, the effects of contact force are not altered by muscle contraction and a likely candidate for the mechanism responsible for contact force artefact is the viscoelastic compression of superficial tissues (skin and adipose) during loading. Simulation data suggests that adipose tissue plays a key role in diffuse reflectance of photons, so any compression of the superficial tissues will affect the reflected signal. Further research is required to fully understand the mechanisms behind contact force artefact, which will, in turn, inform future NIRS device designs.

Recent literature published by some practitioners, consultants, and researchers in the area of Business Process Management (BPM) identified that BPM is a new and emerging field of research and practice. The objective of this paper is to identify the conceptual framework of BPM, identify if connections exist with prior process improvement concepts such as Business Process Re-engineering (BPR), Total Quality Management (TQM) and Business Process Improvement (BPI), and apply BPM in a case study to determine the effectiveness of the current methodology. An extensive literature review was conducted, identifying multiple similarities between BPM and prior process improvement concepts, suggesting an evolving nature of the concept. The BPM methodology was then applied in a controlled case study, identifying a major inefficiency in the methodology. The findings of this paper are useful to researchers, educators, students, and managers to understand the evolution of BPM, and determine how it can be applied.

Recent advancement of artificial intelligence (AI) techniques have impacted the field of algorithmic music composition, and that has been evidenced by live concert performances wherein the audience reportedly often could not tell whether music was composed by machine or by human. Among the various AI techniques, genetic algorithms dominate the field due to their suitability for both creativity and optimization. Many attempts have been made to incorporate rules from traditional music theory to design and automate genetic algorithms. Another popular approach is to incorporate statistical or mathematical measures of fitness. However, these rules and measures are rarely tested for their validity. This thesis is aimed at addressing the above limitation and hence paving the way to advance the field towards composing human-quality music. The basic idea is to look beyond this constrained set of traditional music rules and statistical/mathematical methods towards a more concrete foundation. We look to a field at the intersection of musicology and psychology, referred to as music-psychology. To demonstrate our proposed approach, we implemented a genetic algorithm exclusively using rules found in music-psychology. An online survey was conducted testing the quality of our algorithm’s output compositions. Moreover, algorithm performance was analyzed by experimental study. The initial results are encouraging and warrant further research. The societal implications of our work and other research in the field are also discussed.

As the world is flooded with data, the demand for mining data for useful purposes is increasing. An effective techniques is to model the data as networks (graphs) and then apply graph mining techniques for analysis. As on date, the algorithms available to count graphlets and orbits for various types of graphs and their generalizations are limited. The thesis aims to fill the gap by presenting a simple and efficient algorithm for 3-node graphlet and orbit counting that is generic enough to work for both undirected and directed graphs. Our algorithm is compared with the state-of-art algorithms and we show that in most cases our algorithm performs better. We demonstrate our algorithm in three case studies related to (i) enzyme and metabolite correlation network in corn, (ii) watershed governance networks, and (iii) patterns exhibited by co-expression networks of healthy and cancerous stomach cells.

The design of efficient control strategies is a well studied problem. Due to recent technological advancements and applications in the field of robotics, exploring novel ways to design optimal control for multi-robot systems has gained interest. In this respect, the concept of ergodicity has successfully been applied as an effective control technique for tracking and area coverage. The generation of flocking behaviour is a problem that involves both tracking and coverage, and as such is also suited for the use of ergodicity. The main contribution of this thesis is the application of ergodicity to emulate flocking behaviour. This approach is appealing because control and communication is assumed to be local, self-organized, and does not require separate algorithms in order to generate different behaviour. Simulation results show that the proposed approach is effective and a prototype provides evidence that flocking behaviour is possible using ergodicity in a real-life setting.