UNSW Canberra

Publication Search Results

Now showing 1 - 7 of 7
  • (2022) Purwanto, Rizka
    Thesis
    Despite the availability of toolbars and studies in phishing, the number of phishing attacks has been increasing in the past years. It remains a challenge to develop robust phishing detection systems due to the continuous change of attack models. We attempt to address this by designing an adaptive phishing detection system with the ability to continually learn and detect phishing robustly. In the first work, we demonstrate a systematic way to develop a novel phishing detection approach using compression algorithm. We also propose the use of compression ratio as a novel machine learning feature, which significantly improves machine learning based phishing detection over previous studies. Our proposed method outperforms the use of best-performing HTML-based features in past studies, with a true positive rate of 80.04%. In the following work, we propose a feature-free method using Normalised Compression Distance (NCD), a metric which computes the similarity of two websites by compressing them, eliminating the need to perform any feature extraction. This method examines the HTML of webpages and computes their similarity with known phishing websites. Our approach is feasible to deploy in real systems with a processing time of roughly 0.3 seconds, and significantly outperforms previous methods in detecting phishing websites, with an AUC score of 98.68%, a G-mean score of 94.47%, a high true positive rate (TPR) of around 90%, while maintaining a low false positive rate (FPR) of 0.58%. We also discuss the implication of automation offered by AutoML frameworks towards the role of human experts and data scientists in the domain of phishing detection. Our work investigates whether models that are built using AutoML frameworks can outperform the results achieved by human data scientists in phishing datasets and analyses the relationship between the performances and various data complexity measures. There remain many challenges for building a real-world phishing detection system using AutoML frameworks due to the current support only for supervised classification problems, leading to the need for labelled data, and the inability to update the AutoML-based models incrementally. This indicates that experts with knowledge in the domain of phishing and cybersecurity are still essential in phishing detection.

  • (2022) Yi, Jie
    Thesis
    Arterial stenosis is a problem of immediate significance, as cardiovascular disease is the number one leading cause of death worldwide. Fractional flow reserve (FFR) was proposed to evaluate the functional severity of coronary plaque-induced stenosis more accurately. FFR relies on invasive pressure measurements, while computational fluid dynamics (CFD) studies have been demonstrated to be useful tools to predict FFR less invasively. Myocardial bridging (MB) is an abnormality of the epicardial coronary artery where a segment of artery tunnels through the myocardium. MB presents as a ‘dynamic’ stenosis, in contrast to the ‘fixed’ stenosis caused by plaque: in systole, the artery is compressed due to the heart compression force, while in diastole the compression is non-significant. The objective of the project is to replicate the MB compression phenomenon via fluid-structure interaction (FSI) analysis and identify its impact on FFR. The relationship between ‘fixed’ stenosis and FFR was analyzed as a reference firstly, followed by the introduction of a pressure wire and surface roughness, to determine their impacts on CFD-derived FFR. Secondly, both commercial software and in-house code solver were used to perform FSI study and investigate the mechanism of bridging. With increasing severity of the ‘fixed’ stenosis – 0% to 70% diameter reduction, FFR decreased from 0.96 to 0.55. The presence of the pressure wire led to an overestimation of FFR by 3%-38% in various degrees of stenosis model, while the impact of the surface roughness on FFR was not apparent. Mild MB was studied via COMSOL simulations, while moderate and severe MB models were computed with the in-house code solver. The combination effect of the pressure wire and the upstream plaque in the mild MB was not additive, which was larger than the separate effect caused by each factor. With the increasing of the compression of MB – 44% to 60% diameter reduction, FFR decreased slightly, where the values were larger than 0.92. However, FFR dropped noticeably from 0.84 to 0.75 when the compression of MB increased from 72% to 87%. Furthermore, an expansion was observed in the severe MB model due to a greater inner pressure than outer compression pressure. In conclusion, the flow dynamics of MB were quite different compared to the plaqueinduced or ‘fixed’ stenosis. The use of traditional FFR to evaluate the functional severity of MB should be applied carefully and the cut-off value needs to be amended accordingly.

  • (2023) Varsha Sivagurunathan, Varsha
    Thesis
    The urban water sector is confronted with a multitude of challenges. Rapid population growth, changing political landscapes, aging water infrastructures, and the worsening climate crisis are creating a range of uncertainties in the sector around managing water. Scenarios have been used extensively in the environmental domain to plan for and capture uncertainties to develop plausible futures, including the field of urban water management. Scenarios are key in enabling plans and creating roadmaps to attain desired futures. Despite the advantages and opportunities that scenarios offer for planning, they also have limitations; generally, and within the urban water space. Firstly, the growing uncertainty surrounding urban water management systems necessitates a focused review specifically aimed at the use of scenarios in urban water management. This thesis presents a systematic review to empirically investigate the crucial dimensions of urban water scenarios. Through this review, key knowledge gaps are highlighted, and recommendations are proposed to address these gaps. Secondly, scenarios often depict distressing, almost dystopian futures. Though negative future visions help understand the consequences of present trends and aid in anticipating imminent threats, the limited exploration of positive future visions can make it challenging to find the direction to transform. Optimistic scenarios delve into what people want for the future and capture how their aspirations shape them. Imagining positive visions encourage innovative thinking, creates agency, and creates pathways to desired futures. There is therefore a recognition to move towards more positive, desirable futures. This thesis uses a narrative, participatory scenario process, the SEEDS method, to develop positive visions of urban water futures. The Greater Sydney region in New South Wales, Australia is used as a case study to evaluate the applicability of this approach for urban water management. The urban water sector in the Greater Sydney region faces a multitude of challenges including impacts from climate change, managing diverse water supply sources, and meeting future water demand. These challenges create an increasingly uncertain future for the water sector, where the scale and nature of water services needed in the Greater Sydney region can be unclear. Hence, the Greater Sydney region is selected as the case study region to apply the SEEDS method and develop scenarios for urban water management to plan for future uncertainties. Thirdly, only a few scenario studies include surprises, the unexpected events, which make scenarios useful for planning. Challenges around capturing surprises in scenarios include a lack of structured approaches as well as a lack of evaluation of those methods that have been developed. This thesis discusses the effectiveness and suitability of various surprise methods for scenario development. These methods have been applied in the context of the SEEDS method for urban water management. Finally, there is a lack of evaluation of the tools used to cope with surprises as well as a lack of evaluation efforts of urban water management scenario studies. The assessment of the SEEDS approach for urban water management as well as the different surprise methods for scenario development requires evaluation criteria. This thesis develops and presents an evaluation criteria list based on existing literature that captures key criteria required for adequate assessment of the surprise methods and the scenario process. This thesis contributes to the fields of scenario development and urban water management, and the use of surprises within scenarios. Critical gaps in existing urban water management scenario practices are highlighted and key recommendations are proposed to fill the gaps. Through the pilot study and full-scale implementation of a positive-visioning, narrative-based scenario approach - the SEEDS method, the thesis demonstrates that the SEEDS method is applicable for urban water planning and shows potential for use at different stages of water planning. The positive visions generated through the SEEDS method highlight fundamental aspirations for the urban water sector, possible challenges, and conflicts, and discuss pathways to achieve positive future visions. By using in-situ experimentation and engaging participants with expertise in the relevant field, this thesis provides a realistic evaluation of the scenario process and surprise methods. This thesis thus fills the critical gap about the lack of evaluation in urban water management scenario processes by assessing the scenario method using selected evaluation criteria. Further, the thesis contributes towards the development of quality surprise methods through application and evaluation, thus addressing the gap about the lack of evaluation of the methods used to explore surprise events. Finally, the lack of surprises in scenarios is addressed by presenting different methods that can be used to explore surprise events. Guidance is provided to researchers working with scenario development to understand the different surprise methods available and for choosing the appropriate method(s) to plan for uncertain futures.

  • (2023) Nawaz, Muhammad Haq
    Thesis
    Piezoelectric materials are the preferred smart materials for sensing and actuation in the form of micro and nano-engineering structures like beams and plates. Cantilever beams play a significant role as key components in atomic force microscopy and bio and chemical sensors. Adding an active layer such as lead zirconate titanate (PZT) thin-film to form smart cantilever beams with sensing and actuation capabilities is highly desirable to facilitate miniaturization, enhance performance and functionali- ties such as enabling on-chip high-speed parallel AFM. During the micro-fabrication process, residual stresses develop in the different layers of the cantilever beam, causes initial deflection. The residual stress in the different layers of the cantilever beam and the application of voltage to the PZT thin-film affects their dynamics. This the- sis investigates the dynamic behaviour and develops a control technique and a novel charge readout circuit to improve the performance of a micro-fabricated multi-layer prestressed piezoelectric cantilever beam as an actuator and a deflection sensor. Firstly, the fabrication process of a unimorph PZT cantilever beam is explained. A low thermal budget Ultra-high vacuum e-beam evaporated polysilicon thin-film (UHVEEpoly) process is used for the fabrication of multi-layered PZT cantilever beam in d31 mode. The sharp peaks at resonant frequencies in the frequency response of the PZT cantilever beam show very little damping and a large settling time of the cantilever beam. Secondly, the dynamic behaviour of the prestressed PZT cantilever beam is in- vestigated subjected to change in driving voltage. Experimental investigations show a shift in resonant frequencies of a PZT cantilever beam. However, there is no reported mathematical model that predicts the shift in resonance frequencies of a multi-layered prestressed piezoelectric cantilever beam subjected to a change in driving voltage. This work developed a mathematical model with experimental val- idation to estimate the shift in resonance frequencies of such cantilever beams with the change in the driving voltage. A very good agreement between the model predic- tions and experimental measurements for the frequency response of the cantilever beam at different driving voltages has been obtained. A novel linear formulation has been developed to predict the shift in resonance frequencies of the PZT can- i tilever beam for a wide range of driving voltages. The formulation shows that the shift in resonance frequencies of a multi-layered prestressed piezoelectric cantilever beam per unit of applied voltage is dependent on geometric parameters and material properties. Thirdly, a robust resonant controller has been designed and implemented to re- duce the settling time of a highly vibrating PZT cantilever beam. The controller design is based on a mixed negative-imaginary, passivity, and a small-gain approach. The motivation to design a resonant controller using the above-mentioned analyti- cal framework is its bandpass nature and the use of velocity feedback, as the charge collected from a vibrating PZT cantilever beam gives the velocity information of the beam. The proposed controller design results in finite gain stability for a pos- itive feedback interconnection between two stable linear systems with a large gain and phase margin. Experimental results demonstrate that the designed resonant controller is able to effectively damp the first resonant mode of a cantilever, signifi- cantly reducing settling time from 528 ms to 32 ms. The robustness of the designed resonant controller is tested against changes in the cantilever beam dynamics due to residual stress variation and or stress variation due to driving voltage. Finally, to facilitate the miniaturization of on-chip sensors and parallel high- speed AFM, a single layer of a PZT thin-film in a cantilever beam is used as a deflection sensor and an actuator instead of bulky optical deflection sensors. A novel charge readout circuit is designed for deflection sensing by capturing the electrical charge generated due to the vibration of the PZT beam. The signal-to-noise ratio and sensitivity analysis of the readout circuit shows similar results compared to the commercially available optical deflection sensors. Our work highlights very important aspects in the dynamic behaviour and perfor- mance of a multi-layered prestressed piezoelectric cantilever beam. The agreement between the proposed theoretical formulation and experimental investigations in modeling, control design, and a novel readout circuit will provide the platform for further the development and miniaturization of microcantilever-based technologies, including on-chip parallel HS-AFM.

  • (2023) Rasid, Syed Mamun R
    Thesis
    This thesis presents modeling and control of an on-chip self-sensing novel multi-layered thin-film piezoelectrically-driven micro-lens actuator that has the potential to be used in micro-optic devices. The actuator is made of six unimorph piezoelectric beams connected tangentially with the lens holding platform to hold a PDMS lens and operate in d31 mode. The use of polysilicon instead of single crystal silicon as a passive structural layer allows for manipulation of residual stress to generate out-of-plane displacement for optical zooming. The actuator exhibits intrinsic residual stress due to different thermal expansion coefficients of the films and annealing processing steps during fabrication, and applied driving voltage generates axial stress in the PZT layer. An analytical model of the actuator taking into account residual stress of the film and driving voltage is developed and validated with Finite Element Analysis (FEA) and experiments. The analytical model accurately predicts the dynamic behavior of the actuator and can be applied to other multi-layered piezoelectric microstructures by adjusting boundary conditions. Feedback control improves the dynamic response of the actuator by increasing damping in the resonance mode. A resonant controller designed based on the identified model of the actuator is implemented experimentally, achieving a damping of 21 dB in the first resonance mode and reducing the settling time to 7 ms from the open-loop settling time of 120 ms. An on-chip self-sensing piezoelectric sensor using current feedthrough cancellation instead of voltage is designed, fabricated, and characterized for use in feedback control. The measured dynamic range of the micro-lens actuator and signal-to-noise ratio (SNR) of the on-chip self-sensing piezoelectric sensor are 22 dB and 108 dB, respectively. A passive resonant feedback controller designed based on feedback from an on-chip self-sensing piezoelectric sensor is implemented experimentally on a piezoelectrically driven micro-lens actuator. The closed-loop implementation using the on-chip self-sensing piezoelectric sensor dampened the resonance mode by 20.6 dB, and the settling time for a square wave input was reduced to 10ms from the open-loop 120 ms. The quick settling time achieved by the on-chip self-sensing enables the realization of a fast micro-lens actuator, leading to fast auto-focusing and large optical zooming capabilities in portable micro-optic devices.

  • (2021) Seyfouri, Moein
    Thesis
    Multiferroic BiFe0.5Cr0.5O3 (BFCO) in which ferroelectric and magnetic orders coexist has gained research interest owing to its potential applications, e.g., spintronic and resistive random-access memory. Moreover, multiferroics possess a narrower bandgap compared to typical ferroelectrics, extending their application to photovoltaic devices. In contrast to the conventional semiconductors, the polarization-induced electric field facilitates the photoexcited charge separation, leading to an above-bandgap photovoltage in ferroelectrics. Nevertheless, a long-standing issue is the relatively low absorption of visible light. Thus, it is essential but challenging to tune their bandgap without compromising ferroelectricity. This thesis explores structural phase transition in the epitaxial BFCO films grown on SrRuO3 buffered (001) SrTiO3 substrate via Laser Molecular Beam Epitaxy (LMBE). Reciprocal space mapping result shows strain relaxation mechanism is not solely by the formation of misfit dislocation but also by changing the crystal symmetry, transitioning from tetragonal-like to a monoclinically distorted phase as the thickness increases. The crystallographic evolution is also coupled with bandgap modulation, confirming that BFCO structure and its physical properties are strongly intertwined. Using spectroscopic ellipsometry, the slight redshift of the bandgap distinguishes the absorption process of the T-like BFCO layer from that of monoclinically distorted structure, further confirmed by spectral photocurrent measurement via conductive-atomic force microscopy. The preparation of pure phase BFCO film with a robust polarization is of paramount importance for practical application. Yet, similar to the parental bismuth ferrite, BFCO suffers from poor electrical leakage performance. We report a three-order of magnitude suppression in the leakage current for the BFCO film through judicious adjustment of the growth rate. Scanning probe microscopy (PFM, AFM and c-AFM) results reveal that both microstructure and ferroelectric properties can be tuned by lowering the growth rate, ensuing realization of the room-temperature ferroelectric polarization comparable to the ab-initio predicted value. This thesis provides a facile strategy to tailor the structure-property of epitaxial BFCO film and its functional response for emerging optoelectronic devices.

  • (2024) Sathirasethawong, Chawin
    Thesis
    Inspire by the idea of cataloguing of insects, an automated 3D digitizing system for small objects is introduced in this thesis. Previous existing systems have complicated equipment and time-consuming image acquisition processes which increase the risk of subject damages. As a possible remedy, the light field imaging concept is introduced for small object scanning. Light field imaging has the unique feature of capturing multiple depths of field in one shot, thus reducing acquisition time, and simplifying the required hardware. In small object photogrammetry, the use of a rotary stage decreases the acquisition setting’s complexity. While the setting is simpler, image masking is crucial for the separation between the object and the background. Although there are many masking algorithms in the literature, they have limited performance with 2D images. With the additional depth cue present in light field images, novel object masking algorithms specific to light field images are proposed. The techniques can extract the object of interest from the background by density-based spatial clustering of applications with noise, together with morphological filtering. The result shows that the algorithm works well in varieties of background environment. Moreover, by utilizing the information embedded in the light field images, the algorithms are capable of projecting the mask images from the centre sub-aperture image to adjacent views. Reconstruction of the models is done by implanting photogrammetry. With light field slope information that is obtained from our masking algorithm, the depth specific light field feature extraction algorithm is developed. Without utilising the masks in image preprocessing, the algorithm assigns the depth parameters used in light field feature extraction automatically which speeds up the processing time and also overcomes mistaken features from the background. Throughout this thesis, we introduce the automated 3D digitizing system of small objects using light field technique as a core. The system is flexible and requires less acquisition time for objects that need macrophotography. Moreover, the novel unsupervised object masking algorithm is developed. The developed masking algorithm is promising and helpful especially when there are hundreds of input images. Lastly, the depth specific light field feature extraction algorithm is developed which provides faster processing time and rules out the background even without relying on the masks.