85000 research outputs found
Sort by
Multi-layered laminate architectures enhance the electromechanical response of PVDF–TrFE films
No full textThe digitisation of society and the rise of autonomous sensing networks have led to the need for independent and autonomous low-grade power supplies. Electromechanical harvesters, which convert motion or vibrations into electricity, show great promise for powering microelectronic devices. In particular, poly(vinylidene difluoride (PVDF) is an exemplar polymer material for electromechanical conversion and is highly attractive as a power source due to its exceptional chemical stability. In this study, we present an approach to enhance the electromechanical conversion of a PVDF co-polmyer by electrospinning multi-layered nanofiber laminate architectures, consisting of alternating smaller and larger diameter fibers. This alternative layer structure results in the introduction of oriented triboelectric dipoles within the volume of the laminate architecture, able to couple with the piezoelectric dipole of PVDF. The laminate shows an 11× improvement in electromechanical conversion when compared to an equivalent single diameter fiber network under identical conditions. The application of the laminate architecture to airflow and sound energy harvesting is also demonstrated, with a doubling of the peak-to-peak short-circuit current compared to a state-of-the-art commercial poled PVDF film. This approach provides a pathway to improve the electromechanical performance of PVDF for a wide array of electromechanical conversion applications.</p
Housing For a Changing Climate
No full textThe increase in severity and frequency of climate related impacts has both financial and social implications for the housing sector. Resilient housing is not only about the material conditions that support our climate resilience, but also the resilience of the people who live in those houses. We argue that we must start to deliver new and retrofit housing that will perform well now, and into a changing future.</p
Situated Attentions Listening With the Local in Creative Practice
No full textThis project explores situated attention in and with place as a guide for the development of hybrid listening/browser-based creative works. Three experimental, hand-coded webpages produced in and with the more-than-human neighbourhood of ‘Footscray’ and the inner west of Naarm/‘Melbourne’ on Wurundjeri Country are the focus of the dissertation: Local Time, this time around, and dispersal. Examining associated fieldwork, soundwalking, foraging, gardening, audio recording, listening and experimental creative coding uncovers a narrowing of attention, from ambient to embodied, to interaction enacted with a practitioner’s hands. By paying close attention to the vital role of the attentive body of the maker, while contemplating questions and sitting with tensions of settler-colonial belonging, making-with, the ethics of place-based creative practice on stolen land, and materialities of web infrastructures, this PhD argues that situated knowledges of place are vital to careful, ethical and sustainable practice. These situated knowledges are centred on local attention as a way to refuse, make visible, and create alternatives to default modes of being in place and in the browser. The practice-based research presented here offers an assemblage of contributions to interdisciplinary communities of practice including acoustic ecology, creative audio, web design and net art, and to local more-than-human communities. https://situatedattentions.localtime.place/</p
Next-Generation Collaborative Learning with Integrated Trust, Privacy, Governance and Scalability Solutions
No full textThis thesis investigates novel strategies to enhance trust, privacy, auditability, and scalability in federated learning systems. In distributed environments, the collaborative training of machine learning models faces critical challenges such as adversarial manipulation, privacy breaches, and inconsistencies arising from data heterogeneity and diverse client capabilities. Addressing these issues requires a multifaceted approach that secures model aggregation, verifies individual contributions, and optimizes resource allocation without compromising performance. The initial segment of this research focuses on reinforcing the trustworthiness of the aggregation process. In federated learning, local models are vulnerable to tampering, which can result in a corrupted global model. To mitigate these risks, a secure aggregation procedure is proposed in which each local model is rigorously validated prior to its incorporation into the global model. A consensus mechanism is employed to ensure that the aggregated model adheres to the highest standards of integrity, with every update recorded in a tamper-resistant manner. Experimental evaluations conducted across various neural network architectures and benchmark datasets demonstrate that this method maintains model accuracy while effectively countering potential adversarial threats. Subsequently, the thesis addresses the imperative of privacy preservation in scenarios involving sensitive data, such as those encountered in healthcare and industrial applications. By integrating advanced privacy-preserving techniques, the study facilitates the encrypted processing of local model updates, thereby ensuring that sensitive information remains confidential throughout the training and aggregation processes. Empirical results indicate that these privacy measures introduce only marginal overhead while preserving the efficacy of the learning process, thereby providing a balanced solution that upholds data confidentiality without impeding model performance. Furthermore, the research emphasizes the necessity for auditability and verifiability within federated learning systems. Recognizing that decentralized training environments often lack centralized oversight, a robust governance mechanism is introduced to record detailed training logs and aggregation steps. This verifiable approach enables all participants to independently confirm that the prescribed protocols have been followed, thereby enhancing transparency and accountability. The capacity to audit the entire learning process not only strengthens stakeholder confidence but also aligns with evolving regulatory standards in data protection and machine learning ethics. The final aspect of the study tackles the challenges of scalability and resource optimization. Real-world federated learning applications are characterized by heterogeneous client devices and variable network conditions, which can impede efficient training. To address these issues, dynamic resource management techniques are developed that adaptively partition and allocate computational tasks based on the capabilities of each participant. Comprehensive simulations and real-world experiments reveal that these methods accelerate model convergence, reduce training time, and optimize energy consumption, all while maintaining high levels of global model accuracy. In summary, the contributions of this thesis are demonstrated through a comprehensive approach that fortifies federated learning systems against security threats, ensures privacy preservation, and enhances both auditability and resource efficiency. By integrating secure aggregation methods, privacy-preserving processing, verifiable governance, and dynamic resource management, this research offers a robust solution to many of the pressing challenges in distributed machine learning. The findings provide a solid foundation for future investigations aimed at further refining consensus protocols and extending the approach to accommodate heterogeneous model structures in increasingly dynamic environments.</p
Workplace security and privacy implications in the GenAI age: A survey
No full textGenerative Artificial Intelligence (GenAI) is transforming the workplace, but its adoption introduces significant risks to data security and privacy. Recent incidents underscore the urgency of addressing these issues. This comprehensive survey investigates the implications of GenAI integration in workplaces, focusing on its impact on organizational operations and security. We analyze vulnerabilities within GenAI systems, threats they face, and repercussions of AI-driven workplace monitoring. By examining diverse attack vectors like model attacks and automated cyberattacks, we expose their potential to undermine data integrity and privacy. Unlike previous works, this survey specifically focuses on the security and privacy implications of GenAI within workplace settings, addressing issues like employee monitoring, deepfakes, and regulatory compliance. We delve into emerging threats during model training and usage phases, proposing countermeasures such as differential privacy for training data and robust authentication for access control. Additionally, we provide a comprehensive analysis of evolving regulatory frameworks governing AI tools globally. Based on our comprehensive analysis, we propose targeted recommendations for future research and policy-making to promote responsible and secure adoption of GenAI in the workplace, such as incentivizing the development of explainable AI (XAI) and establishing clear guidelines for ethical data usage. This survey equips stakeholders with a comprehensive understanding of GenAI's complex workplace landscape, empowering them to harness its benefits responsibly while mitigating risks.</p
Producers’ Perspectives on Australian Children’s Television in the Streaming Era
No full textThis report presents findings from semi-structured interviews conducted with 23 producers of Australian children’s television, conducted between November 2021 and April 2024, which was a period of intense legislative, technological, and funding changes. The report gathers and analyses producers’ expert perspectives on the sector, to map global trends, opportunities and challenges facing Australian children’s television producers.</p
Manifesting Landscape: A Painterly Approach to Virtual Reality Creation
No full textThis PhD investigates the creative and artistic potential of head-mounted virtual reality (VR) technology as a generative medium, focusing on the manifestation of landscapes within digital virtual environments. Drawing on a painterly approach informed by my past practices in landscape painting, this PhD emphasises the dynamic interaction between artist-technology and the physical-digital. This research speculates landscape as a unique spatial paradigm in digital environments, taking influences from the philosophy and aesthetic tendencies of Xie Yi in traditional Chinese landscape paintings. Further to this, I am interested in expanding the artistic and practical potentials of VR landscape practices by foregrounding the expressions of bodily movements. The research is guided by the central question: ‘How to manifest landscapes with(in) VR through VR creation?’ Here, VR creation is defined as an emerging practice of creating in-VR-for-VR, while the term ‘manifesting’ is used to denote both the digital modulation of VR and the conceptual process of interpreting and expressing these creations. Manifesting VR landscapes involves the active interplay of physical and conceptual processes to actualise both material form and meaning. This process transcends the re-presentation of natural landscapes in a digital virtual environment, establishing a more intimate relationship between the creator and their creations. Consequently, the definition of ‘landscape’ in this research is grounded by personal memories and imaginations. Through the practice research of VR explorations, reflections, and lived experiences, this PhD illuminates the more sentimental qualities of VR creation as an emerging practice for manifesting landscapes. The research began as an inquiry into the creative parameters of VR technology and has since expanded the definition of landscape by critically engaging with these parameters. Through a series of formal studies, I explored the inherent features of VR and the symbiotic relationship between dynamic human input and technological mediation. I recognised how bodily gestures and motions become dynamic inputs that makes VR creation a unique digital practice; I leveraged the technical capacities of VR and developed various techniques for generating and transforming VR forms. These techniques are inspired by and revised to physical painterly mediums, including my practice of oil painting and charcoal drawings and the brush and ink techniques from Chinese landscape paintings. These explorations problematise the spatial paradigms of landscapes in digital virtual environments, expanding on painterly practices of landscapes such as traditional Chinese landscape paintings and plein air oil paintings by probing the process of manifesting VR landscapes. A key turning point of the practice research occurred when Xie Yi techniques became a significant influence on my painterly approach to manifesting VR landscapes. Xie Yi is a Chinese painting tradition rooted in traditional Chinese philosophy and aesthetics. It emphasises the elusive nature of expression, prioritising spiritual essence over figurative depiction (Wu, 2020). Practising VR creation through the lens of Xie Yi deepened my understanding of bodily gestures and the intuitive processes of actualising impressions of landscapes in visual forms. This has enabled the development of a painterly approach distinct from conventional 3D modelling for creating VR elements. Chinese landscape paintings and VR creation, despite the difference in medium, there is a shared emphasis on the dynamic processes that explore the indeterminacies of landscapes. These indeterminacies refer to their fluid forms, boundaries and spatial paradigms actualised through the medium relative to the creator. Through this process, I have examined VR not only as a technological tool but also as a digital virtual site and creative medium that the creator actively influences and is influenced by. The practice research has highlighted the potential of VR landscape practices while also recognising the limitations of the current VR technology as a creative medium. Practising through the lens of Xie Yi helped articulate the multifaceted process of VR creation with attention to bodily expressions. I then conceptualised and tested a multi-dynamic gestural input system for VR painting. This system integrates gesture recognition and pressure sensing to enhance the expressive potential of VR. It has enabled me to manifest VR landscapes more intuitively, attuning to the qualities of VR as digital and virtual, and qualities of landscapes as dynamic and indeterminate. By correlating sensory inputs with the formal qualities of VR lines—size and opacity—the system has allowed me to incorporate more nuanced expressions in VR creation that are akin to Chinese brush and ink techniques. In foregrounding the bodily and hand gestures in VR creation, dynamic expressions of VR forms coalesce into a unified VR ‘land-scape’, in which the fragmented VR lines are regarded as the production of continuous motions. In this research, VR and landscape became intertwined, representing an emerging practice native to the digital virtual. This dual role—as medium and environment—inspired the term land-scape. I derived the term from the Dutch word landschap, meaning 'a piece of land’ (Grout, 2000, p.1). This PhD expands the potential of VR creation as a painterly practice that redefines landscape within digital virtual environments, therefore contributing to both the theoretical and practical understanding of VR landscapes. It probes the creative possibilities of VR as an emerging artistic practice while addressing broader conceptual questions of space and bodily expressions in digital virtual environments. The practice research of manifesting VR landscape through a painterly approach further highlighted the qualities of VR creation as a digital virtual and spatial practice. In problematising VR landscape and VR creation, I have derived approaches, creative techniques, and interactive systems that suggest ways of harnessing VR’s attributes for VR creation.</p
Storytelling and Brand Advertisement: An Analysis of TikTok Influencers in Vietnam
No full textTikTok has become a phenomenon when it attracts a big number of followers thanks to its creative video content and the users who make it, so brands choose TikTok as a channel of communication to endorse their brands in the users’ videos via many narrative styles. Using the qualitative research with case study approach, the paper aims to cover three main points: (1) common storytelling techniques, (2) influencers’ advertisement videos using these techniques and purchase intentions, and (3) micro influencers used for advertising campaigns on TikTok. The finding shows that the most common storytelling techniques on TikTok are humorous content, product review, storytelling dance, and making an experience relating to products and life; TikTok platform is used for influencer marketing in which micro/ nano influencers are hired to endorse a brand in their videos for advertisement via experience content and continuous engagement impacting their followers on purchase intention; the choice of influencers depends on a brand’s budget and communication objectives besides other criteria but a mix of different types of influencers is more feasible. The research implicates the potential trend in using influencers in TikTok to create brand advertisement content with various storytelling methods in a brand advertising campaign in recent years.</p
Hexagonal Ge on Self-assisted GaAs Nanowires by MBE
No full textThe increasing demand for data requires photonic technologies to overcome the limitations of traditional electronic interconnects. Silicon photonics offers a solution, but the integration of efficient light sources remains challenging due to silicon’s indirect bandgap. Germanium (Ge), though also indirect in its cubic phase, can achieve a direct bandgap in its hexagonal phase (h-Ge), making it a promising candidate for light emission.
This thesis focuses on growing h-Ge on self-assisted GaAs nanowires (NWs) using molecular beam epitaxy (MBE). Optimized growth of GaAs NWs in the Wurtzite (WZ) phase enables high-quality h-Ge deposition. Structural and optical characterization confirm the presence of hexagonal Ge and demonstrate tunable light emission via quantum confinement, moving closer to telecommunication wavelengths. Challenges such as defects and contamination are discussed. This thesis highlights the potential of h-Ge for next-generation photonic devices.</p
The Resilience of Tradition: Capturing and Examining the Role Of the Traditional Economy in Fiji
No full textTo date, development initiatives in developing countries, such as the small Pacific Island nation of Fiji, have failed to increase broad wealth and prosperity (Thompson 2001; UN 2014). These nations often have prevalent traditional economies – i.e., economies where unique customary roots define economic behaviour. The prevailing economic ideology of development has been that developing countries will gradually transition from their traditional economies to a formal economy through the process of economic growth as measured by rising Gross Domestic Product. However, these economies have struggled to sustain meaningful growth, are reliant on foreign assistance and increasingly financially liable to foreign interests. Moreover, the traditional economy can be a crucial source of economic support and community wellbeing, especially in times of economic shock, which means that disassembling it can be counterproductive to development. From the United Nations Sustainable Development Goals through to local governments in the Pacific, there is increasing recognition that development agendas should recognise the productivity of traditional economies and not base economic growth solely on GDP. There have been no economic studies, to the best of my knowledge, that have sought to measure traditional economies and assess their role in economic growth and wellbeing. Addressing this gap, this thesis seeks to capture and investigate the significance of the traditional economy in the case study of Fiji and examine its role in meaningful development. The findings from the research outline that the traditional economy is steadfast in Fiji and is important for wellbeing. The thesis proposes greater state support and enhanced technology for the agricultural sector to support and commercialise the traditional economy which, alongside a robust engagement with the globalised formal economy, can foster meaningful and sustainable economic growth.</p