AUT Research Repository (Auckland Univ. of Technology)
Not a member yet
14707 research outputs found
Sort by
Joint Effect of Signal Strength, Bitrate, and Topology on Video Playback Delays of 802.11ax Gigabit Wi-Fi
No full textThis paper presents a performance evaluation of IEEE 802.11ax (Wi-Fi 6) networks using a combination of real-world testbed measurements and simulation-based analysis. The paper investigates the combined effect of received signal strength (RSSI), application bitrate, and network topology on video playback delays of 802.11ax. The effect of frequency band and client density on system performance is also investigated. Testbed measurements and field experiments were conducted in indoor environments using dual-band (2.4 GHz and 5 GHz) ad hoc and infrastructure network configurations. OMNeT++ based simulations are conducted to explore scalability by increasing the number of wireless clients. The results obtained show that the infrastructure-based deployments provide more stable video playback than the ad hoc network, particularly under varying RSSI conditions. While the 5 GHz band delivers higher throughput at a short range, the 2.4 GHz band offers improved coverage at reduced system performance. The simulation results further demonstrate significant degradation in throughput and latency as client density increases. To contextualize the observed performance, a baseline comparison with 802.11ac is incorporated, highlighting the relative improvements and remaining limitations of 802.11ax within the evaluated signal and load conditions. The findings provide practical deployment insights for video-centric wireless networks and inform the optimization of next-generation Wi-Fi deployments
Experienced Well-Being and Compliance Behaviour: New Applications of Quality of Life theories, Using AI and RealTime Data
No full textThe study of well-being has continued to evolve significantly over the past three decades, extending the foundational progress documented by Diener et al. (1999) through advances in measurement, cross-national surveys, and the emergence of high-frequency, real-time indicators. One of the most pressing issues in contemporary well-being research is the intersection between experienced well-being measures and societal compliance, especially in times of uncertainty. Effective crisis response depends not only on well-designed policies but also on how populations emotionally interpret uncertainty and respond behaviourally. This paper introduces a framework in which experienced well-being indicators are repositioned as behavioural inputs that shape compliance with public health interventions. Drawing on interdisciplinary theories, we argue that emotional readiness plays a critical role in driving prosocial behaviour during times of crisis. Using a macro-panel at the country–day level dataset and applying XGBoost and SHAP, we examine how dynamic, within-country features, both structural and subjective, predict compliance with COVID-19 vaccination policy. Results show that general trust and happiness are among the strongest predictors of compliance, often rivalling or exceeding traditional factors like GDP per capita or healthcare spending. Our findings show experienced well-being indicators not only predict compliance within countries but also have cross-national relevance, providing a foundation for more psychologically informed policy design. We propose that policymakers integrate these emotional indicators into crisis response systems to improve behavioural effectiveness and public cooperation
From Immersion to Identity: A Systematic Review and Framework for Understanding Metaverse Consumers
No full textMetaverse marketing has emerged as a rapidly expanding research domain, yet scholarship on consumer behaviour within this context remains fragmented and often embedded within broader marketing or technology reviews. This study addresses this gap through a systematic literature review (SLR) of 84 peer-reviewed journal articles explicitly situated at the intersection of consumer behaviour and marketing in the Metaverse. The review synthesises insights into four thematic domains: (1). consumer engagement, (2). technology adoption, (3). avatar dynamics, and (4). fashion/luxury branding. The review maps three theoretical foundations: motivational, identity-based, and design-oriented perspectives. By integrating these strands, the paper develops a novel conceptual framework that captures how immersive, multisensory, and identity-driven experiences in the metaverse shape consumer–brand interactions and extend into real-world consumption. The analysis highlights the need for metaverse-specific behavioural theorisation and outlines key research gaps, offering a targeted agenda to advance understanding in this evolving field
Remaking the Imago Paterna: A Heuristic Enquiry
No full textJung’s imago encapsulates mythic, internal representations of others, born from subjective experience, culture, and archetypes. Often, psychotherapy analyzes and reconstructs this image. In this article, heuristics were used to explore the first author’s image of his father—assembled from fragmented subjectivity and absence. Following Moustakas’s phases, he engaged archetypal themes, literature, and poetry to confront and reimagine this image. Viewed as a mythic journey, psychotherapy becomes both a creative act and a transgenerational rite of passage. The rewriting of imago through therapy reimagines relationships and narratives, complicating Western ideals of individuality while reframing familial and archetypal connections
Lateralized Learning for Multi-class Visual Classification Tasks
No full textThe majority of computer vision algorithms fail to find higher-order (abstract) patterns in an image so they are not robust against adversarial attacks. Deep learning considers each input pixel in a homogeneous manner such that different parts of a locality-sensitive hashing table are often not connected, meaning higher-order patterns are not discovered. Hence, these systems are sensitive to noisy, irrelevant, and redundant data, leading to wrong predictions with high confidence. Adversarial attacks exploit this vulnerability by generating deceptive inputs that mislead AI systems. In contrast, human vision is rarely susceptible to adversarial attacks. Vertebrate brains afford heterogeneous knowledge representation through lateralization, enabling modular learning at different levels of abstraction. This work aims to verify the effectiveness, scalability, and robustness of a lateralized approach to real-world problems that contain noisy, irrelevant, and redundant data. Two well-known and widely used adversarial attacks, the Fast Gradient Sign Method and the Iterative Adversarial Technique, are applied to generate corrupted test images. The experimental results on multi-class (200 classes) image classification tasks demonstrate that the proposed system effectively captures hierarchical knowledge representations, enhancing its robustness. Crucially, the lateralized system outperformed four state-of-the-art deep learning-based systems for the classification of normal and adversarial images by 19.05% − 41.02% and 1.36% − 49.22%, respectively
The Coercive Edge of Kindness: A Critical Analysis of 'Random Acts' in Nursing
No full textKindness is frequently framed as an unassailable virtue, celebrated across social, professional and political domains as a simple and uncomplicated good. It is rarely problematised, and its assumed benefits are seldom interrogated, leaving kindness largely positioned as a self-evident moral imperative. In this paper, we adopt a Foucauldian lens, not to seek an essential definition of kindness, but to consider how it circulates and operates discursively, what effects it produces and what is surrendered in its performance. We position kindness as a discourse that does not merely encourage compassion or generosity but also regulate behaviour, shapes subjectivities and establishes boundaries around what may or may not be said. Through such mechanisms, the imperative to 'be kind' can act to silence resistance, temper critique and foster compliance, functioning as a subtle technology of governance. By problematising kindness in this way, we reveal how a practice so often presented as wholly benevolent can also operate as a powerful disciplinary force. We suggest that alternatives to the disciplinary framing of kindness may be found within First Nations knowledge systems, which offer different ways of understanding generosity and care beyond Western institutional logics. Our purpose is not to argue for the abandonment of kindness, but to highlight that it should not be accepted uncritically; its operations and consequences must be understood in order for it to be engaged ethically and politically
AI, Journalism and News Media in Aotearoa New Zealand
No full textThis 2026 baseline report on artificial intelligence (AI), journalism, and news media in Aotearoa New Zealand is the first of its kind produced and published by the AUT Journalism, Media and Democracy research centre (JMAD). The report offers a glimpse into New Zealand’s AI-assisted digital news media landscape. The author acknowledges that AI tools, principles, and guidelines are rapidly evolving as new technological tools and services are employed by news media organisations.
The report considers some values and risks of AI to news media and journalism, and evaluates ethical and legal issues arising from AI usage. It also explores how AI is used in New Zealand newsrooms, offering some case studies. The report suggests some key policy areas that should be addressed.AUT Journalism, Media and Democracy research centr
Donor Conception and Psychosocial Support Provisions Across Jurisdictions - What's Out There?
No full textAs demand for donor conception (DC) rises the landscape is becoming increasingly complex. DC-linking now occurs through various means, including direct-to-consumer DNA testing, which may reveal DC where this has not been disclosed and make those genetically related known to each other, including earlier than is possible through identity-release provisions in many jurisdictions. Early contact between donors and recipient parents, as well as same-donor siblings is becoming more common. Large sibling groups within and across jurisdictions are increasingly being identified and there is also growing reliance on imported gametes and online donor recruitment platforms. These developments can be associated with challenges for donor-conceived people (DCP), parents, donors and their families, and have led to calls for more accessible and responsive psycho-social support services. This paper maps the DC context in ten Western countries, including the availability of psychosocial support and counselling. Given the growing complexity of DC and its lifelong impact on all involved, we pay particular attention to post-donation counselling support related to disclosure, long-term psychosocial wellbeing, and DC-linking. We identify key challenges in existing DC provisions and support systems and propose improvements that support DCP, donors, parents, siblings, and their families in managing the longer-term implications of DC
Authoritarian Versus Benevolent Leadership Styles: A Moderated Mediation Model of Paternalistic Leadership, Engagement, Job Status and Hospitality Employee Service Performance
No full textThis study examined whether work engagement mediated the impact of paternalistic leadership styles on the service performance of hospitality employees and further investigated if job status (full-time vs. part-time) moderated the impact of paternalistic leadership styles, based on the affective event and partial inclusion theories. Through an analysis of matching data from 286 restaurant employees and their 2129 customers in Thailand, the study found that work engagement mediated the effect of authoritarian leadership, a dimension of paternalistic leadership (father-like) on service interaction quality rated by customers, and that the mediation effect was stronger for full-time employees than for their part-time counterparts. However, the effect of benevolent leadership, the other dimension of paternalistic leadership (mother-like) was neither mediated by work engagement, nor moderated by job status. Theoretical and managerial implications of the findings are discussed for hospitality researchers and practitioners
Generative AI-Enhanced Robust Semantic Communication Architecture for UAV Image Transmission
No full textUnmanned aerial vehicle (UAV) wireless image transmission has gained widespread application across various fields due to its flexibility, yet it faces critical challenges such as resource constraints and degradation of reconstruction quality caused by harsh channel conditions. To address these issues, we designed a lightweight semantic communication backbone network that substantially reduces the computational and storage overhead of UAVs through codebook assistance and efficient encoder-decoder design. On this basis, to tackle severe image degradation under adverse channel conditions, we introduced a generative artificial intelligence-based (GAI) enhancement module. Specifically, we developed a semantic refinement network (SRN) that employs an innovative signal-to-noise ratio (SNR) adaptive feature-wise linear modulation (FiLM) layer to dynamically adjust its refinement strategy based on real-time channel quality, fundamentally transforming the image reconstruction paradigm from traditional signal recovery to conditional content generation. Extensive experimental results demonstrate that our proposed framework significantly outperforms the current state-of-the-art method under extreme channel conditions, highlighting its great potential for achieving robust UAV image transmission in challenging operational environments