13463 research outputs found
Sort by
SDN-Based Edge Computing in Vehicular Communication Networks: A Survey of Existing Approaches
Full text linkImproved communication performance in vehicular communication networks (VCN) is achieved by enabling additional computation and storage support at the levels of edge, fog, and cloud. Specifically, the edge/fog servers located near vehicles enable improved networking performance, especially in terms of load balancing and end-to-end delay. Recently, Software Defined Networking (SDN) in general and SDN-based edge/fog computing in particular have been employed in the context of VCN, enabling further communications performance improvements to be achieved with the help of SDN-based fog/edge computing in the VCN (SDV-F). This paper surveys a large number of innovative approaches proposed for communications performance enhancement in the context of SDV-F and categorizes them based on different aspects, including Artificial Intelligence (AI) and security support. The advantages and limitations of the proposed approaches in each category are highlighted. Finally, various directions that can be considered as promising research avenues by researchers in the future are presented and discussed
Transport Transitions: Advancing Sustainable and Inclusive Mobility Proceedings of the 10th TRA Conference, 2024, Dublin, Ireland - Volume 3: Eco-efficient Mobility Systems
Full text linkThe book series Lecture Notes in Mobility (LNMOB) reports on innovative, peer-reviewed research and developments in intelligent, connected and sustainable transportation systems of the future. It covers technological advances, research, developments and applications, as well as business models, management systems and policy implementation relating to: zero-emission, electric and energy-efficient vehicles; alternative and optimized powertrains; vehicle automation and cooperation; clean, user-centric and on-demand transport systems; shared mobility services and intermodal hubs; energy, data and communication infrastructure for transportation; and micromobility and soft urban modes, among other topics. The series gives a special emphasis to sustainable, seamless and inclusive transformation strategies and covers both traditional and any new transportation modes for passengers and goods. Cutting-edge findings from public research funding programs in Europe, America and Asia do represent an important
source of content for this series. PhD thesis of exceptional value may also be considered for publication. Supervised by a scientific advisory board of world-leading scholars and professionals, the Lecture Notes in Mobility are intended to offer an authoritative and comprehensive source of information on the latest transportation technology and mobility trends to an audience of researchers, practitioners, policymakers, and advancedlevel students, and a multidisciplinary platform fostering the exchange of ideas and collaboration between the different groups
Positive Discoveries: Identity Development and the Experiences of Gifted LGBTQ+ Students in Ireland
Full text linkThis article outlines findings from a mixed methods study, which aims to address this lacuna in the field of gifted education research, by providing insight into the experiences of gifted LGBTQ+ (lesbian, gay, bisexual, transgender, queer) young people, in Ireland. Participants were asked the same set of questions regarding two sites; post-primary school and a gifted summer program. Over three methods of data collection (anonymous questionnaire, focus groups and interviews) the experiences of gifted LGBTQ+ adolescents (N = 120) were explored. The mean age of participants was 18.4. Using thematic analysis of the qualitative data, four superordinate themes were identified, all of which related to identity development for gifted LGBTQ+ adolescents. The study’s key finding across both the quantitative and qualitative data was that peers and staff at a gifted enrichment program had a more positive effect on identity development than their school-based equivalent. This study was guided by the transformative paradigm, which places central importance on the lives and experiences of marginalized communities, uses transformative theory to develop the inquiry approach and links results of social inquiry to action. As the first set of data on the experiences of gifted LGBTQ+ students in Ireland and one of the first of its kind outside the United States, it is hoped that findings from this research will benefit all educators working with this student population while expanding both the field of gifted education research and research on the experiences of LGBTQ+ adolescents
The teaching and learning of reading in an immersion setting: A focus on word recognition
Full text linkWord recognition is vital in the reading process and word reading
relates to comprehension. In immersion settings, there is a lack of
empirical evidence internationally of how the teaching of reading
affects pupils’ literacy development, and what skills and strategies
are employed by pupils in their reading. In Ireland, children learn
to read in Irish and English in Irish language immersion schools.
In recent years, phonics programmes in Irish have been
introduced to schools. But little is known about taught
application of strategies when reading, or if these strategies are
effective. The current study was designed to ascertain how Irish
immersion pupils aged 9–11 recognise words in their reading in
Irish. Assessment tools that focus on word recognition strategies
were developed specifically for this study. Pupils from two
schools (n = 159) in two jurisdictions were selected for
assessments. This study highlights the need for appropriate
assessment tools when assessing reading in two or more
languages and in a minority language context. Teacher
knowledge is vital, and with a greater focus on the orthography
of Irish, teachers could be equipped to assist pupils with
phonemic strategies as well as provide more opportunities to read
Interactive Monuments in the Digitally Mediated City: Examining the Participatory Potential of Virtual Objects in Physical Space
No full textThis practice-based research project examines the participatory potential of digital 3D objects in urban space and tests their role as intangible, interactive artefacts that are capable of reconfiguring urban encounters. Through the process of remediating photographic archive material of contested artefacts (colonial monuments) into a digital 3D form, the project creates and visualises transitory states of removed monuments to explore modalities for embodied, spatial and participatory experiences in urban space utilising augmented reality (AR). The research builds upon methods from the artist’s personal practice, and employs an artefact-led practice-based process and an ethnographic approach, including participant observation, interviews, and ‘mattering’ interventions to evaluate reflective engagement.
The final iteration of the practical component of the project was showcased at the Irish Museum of Modern Art (IMMA), between the 16th and 30th of June, 2023, and addressed both the past (colonial) and present (decolonial) contexts to give participants an opportunity to express their views towards the changing role of monuments. It aimed to connect individuals through shared urban histories and enable reflection on the role of digitally augmented experiences of urban space and their future potential as shared cultural environments.
Through the collection of qualitative data from participant interaction with the digital 3D objects, the research analyses the ways in which the objects reconfigure the spatial and social dynamics of the urban environment in which they are situated. Through the qualitative evaluation with participants, the research demonstrates how a practice-based participatory project can generate awareness and desire for interactive encounters with obscured histories in public space. It also reveals how digital 3D objects can facilitate critical reflection on contested historical narratives and enable participants to contribute to discourses on the implementation of experimental digital technologies towards creating participatory, innovative, reflective and critically engaging public art experiences in urban space
Light Sources and Materials for EUV Lithography
Full text linkThe field of semiconductor manufacturing is heavily dependent on the process of photolithography. Photoresists undergo chemical changes when exposed to light, allowing for patterning of the silicon before other processing steps such as etching and ion implantation are performed. The current state of the art technology is Extreme Ultraviolet lithography. Light-matter interactions are critical to this field. They are exploited in two areas: the source of the EUV light, and in photoresist materials for the lithography itself. Currently, a tin laser produced plasma is used by industry as the light source, but in the past Free Electron Lasers were considered. Light-matter interactions, relevant to both fields, were studied in this project.
A laser produced plasma based EUV light source was built in order to study the spatial and temporal characteristics of the light and plasma. The differences in using tungsten and tin as the source material were investigated. Spectral intensities on the order of 10^13 photons s^−1 nm^−1 sr^−1 at a wavelength of 13.6 nm (photon energy= 91.2 eV) were achieved with both materials. The interaction of intense EUV light with matter was also investigated through the analysis of multiphoton ionisation of neon at the EUVL photon energy of 93 eV, previously recorded at the FLASH FEL. Peak intensities on the order of 10^16 W.cm^−2 allowed for the detection of sequential one-plus-two photon double ionisation.
EUVL resist candidates (e.g., nanoparticle and metal-inorganic based) demand strong EUV absorbers, namely metals. Relaxation dynamics of chromium-oxalate coordination compounds, previously reported as potential EUV resist candidates, were investigated with the aid of low temperature phosphorescence and time-resolved infrared spectroscopy. Excited state lifetimes ranging from milliseconds to picoseconds were revealed
On-demand Crowdsourced Federated Learning over Edge Devices
Full text linkFederated learning (FL) is attributed to training a machine learning (ML) model over a number of distributed devices while keeping all their training data localized. Under these settings, the edge devices perform computations on their local data before sending the required updates to the central server to improve the global model. This approach has
shown great potential since hundreds of devices can potentially contribute to learning a single task without sharing their local data. Despite its success in many domains, current FL systems face significant challenges in scaling client participation and handling data
heterogeneity, which impede the training of high-performing models.
To address these limitations, this thesis argues that there’s a need for a dynamic learning platform where edge devices could volunteer to collaboratively learn a task through FL. It further proposes FedOnDemand, an on-demand crowdsourced FL framework that
dynamically incorporates edge devices based on demand and availability. The research aims to optimize client participation and resource allocation in FL systems to ensure an efficient and scalable learning process.
We present a novel client selection mechanism designed to optimize the contribution of clients based on their computational resources and data quality. By implementing a multi-criterion client selection protocol, the system dynamically selects clients based on their suitability for a given FL task. To secure this process, we incorporate attributebased access control measures, ensuring that client selection is both effective and secure. This approach not only enhances the quality of the contributions but also safeguards the integrity of the FL process. To manage data heterogeneity and improve model robustness, we model FL as a
Bayesian process. Clients employ Stochastic Variational Inference (SVI) to approximate local posterior distributions, while the server utilizes Bayesian learning techniques to aggregate these updates, effectively managing uncertainty. Furthermore, we explore fairnessaware incentive mechanisms based on data valuation, ensuring clients are rewarded proportionally to their contributions. These mechanisms are designed to foster active and robust participation across diverse network environments. Empirical evaluations using benchmark datasets demonstrate significant improvements in convergence speed, model
accuracy, and system scalability compared to traditional FL approaches.
This research contributes to the field by providing a framework that enhances the operational efficiency of FL models and ensures greater participant engagement and system integrity. The implications of this study are far-reaching, potentially influencing future designs of ML systems that require decentralized data inputs across highly dynamic and
privacy-sensitive environments
Understanding Videos by Learning Structured, Robust and Efficient Representations
Full text linkWith an enormous volume of unstructured video content constantly being generated online, designing intelligent systems for automatic understanding of visual data could have a direct and beneficial effect on several fields such as real-world surveillance, robotics, healthcare, entertainment, content retrieval etc. However, extracting meaningful and relevant information from videos still remains a challenging task and an open area of research. Learning powerful representations in the
video domain involves multiple facets such as structural feature learning, modeling motion, multi-modal feature learning, feature disentanglement etc., with the primary goal of holistic video understanding. Recently, self-supervised learning has gained prominence as an effective paradigm for representation learning in images and videos, eliminating the need for additional label annotations. The objective of this thesis is to thoroughly investigate various video modeling techniques, primarily aimed at learning structured, robust, and efficient video representations within the
framework of self-supervised learning. To focus on learning structured video representations, this work first addresses the task of generic event boundary detection by revisiting a self-supervised method
and enhancing it by incorporating a differentiable motion estimation module to capture the generic spatial and temporal diversities in the videos. Extensive experiments on the Kinetics-GEBD and TAPOS datasets demonstrate the efficacy of the proposed approach compared to the other self-supervised state-of-the-art methods. In order to embed robustness into learned video representations, the thesis then
tackles the problem of video anomaly detection from the perspective of recognizing out of distribution samples. A novel method is proposed to generate spatio-temporal pseudo-anomalies by inpainting masked image regions with a pre-trained Latent Diffusion Model and perturbing optical flow using mixup to simulate spatio-temporal distortions. Additionally, a unified framework is introduced to detect real-world anomalies under the one-class classification setting by learning three anomaly indicators: reconstruction quality, temporal irregularity, and semantic inconsistency. Rigorous evaluations on Ped2, Avenue, ShanghaiTech, and UBnormal benchmarks highlight the method’s effectiveness compared to existing state-of-the-art approaches. To learn video representations efficiently, this research proposes a novel and generalizable Trajectory-Aware Adaptive Token Sampler (TATS) module that learns to adaptively sample motion-centric tokens for masked autoencoder (MAE) pretraining by modeling their motion trajectories in videos. Additionally, a unified training recipe is also introduced that facilitates the joint optimization of both
MAE and TATS from scratch using Proximal Policy Optimization to ensure stable convergence during pre-training even with aggressive masking. Comprehensive evaluation on benchmark datasets (Kinetics-400, Something-Something v2, UCF101, HMDB51) for action recognition demonstrates the effectiveness, generalization, transferability, and efficiency of our work compared to the state-of-the-art methods
Understanding Student Learning Behaviours in E-Learning: Insights from STEM and Social Science Modules
Full text linkThe concept of E-learning has gained popularity among universities and students in recent years as E-learning education platforms can record student learning behaviours in its many forms to recognise and analyse student learning styles. However, it is known that the challenges this brings in monitoring student engagement with course material can be considerable and variable between STEM (Science, Technology, Engineering, Math) and Social Sciences courses. This paper applies a graph-based community detection method that integrates the cumulative actions of a student with the Virtual Learning Environment (VLE), through a preprocessing technique, facilitating deeper analysis of student performance using the OULAD dataset1. Our findings reveal that this
method is trustworthy, and we show that it outperforms traditional classification and clustering methods and achieves superior accuracy in evaluating and predicting academic outcomes—encompassing both formative assignments and terminal assessments. Moreover, this method uncovers variations in learning styles among students in STEM and Social Sciences, indicating commonalities and diversity of the
learning approach for the different types of classes, in the same E-learning system
Irish Clinicians' self efficacy in concussion-care within sporting environments from education to practice
Full text linkIntroduction
Concussion-related clinical practices lack quality and consistency worldwide. The level of clinicians’ self-efficacy, a belief in ability to succeed in the context of concussion-care, might be a contributing factor. This research aimed to explore Irish clinicians’ concussion-related self-efficacy during professional education and clinical practice, through the lens of the triadic reciprocal determinism (TRD) model and the four general self-efficacy sources.
Methods
In total, 285 clinicians (certified athletic therapists, chartered physiotherapists, emergency medical services professionals) and 98 final-year athletic therapy (AT) students participated in a cross-sectional, quantitative evaluation of self-efficacy in concussion assessment and management skills, clinical practices and factors impacting self-efficacy. Following that, guided by an interpretative phenomenology approach, qualitative investigation of factors influencing development of self-efficacy during professional education and clinical practice took place. The perceptions of 12 clinicians and 20 AT students were investigated using a semi-structured interview/focus group format, and reflexive thematic analysis approach. Findings of the aforementioned investigations guided development of recommendations for educational practice.
Results
The overall levels of concussion-related self-efficacy among Irish clinicians and AT students were moderate. However, their skill-specific scores varied from very low to high and correlated with frequency of their use with concussed patients. All general sources of self-efficacy were found relevant in the context of concussion-care. Although practice in a relaxed, real-life environment and educators’ feedback were the highest-rated influencing factors, broader environmental and personal factors modified the outcomes of those experiences, in line with the TRD model. Concussion-related self-efficacy is a dynamic attribute that fluctuates throughout clinicians’ professional life. The period of enrolment on professional education programmes is critical for the development of a strong self-efficacy foundation.
Conclusion
Concussed patients in Ireland receive care from clinicians who feel only moderately efficacious about delivery of optimal concussion-care. Professional and personal development should be promoted throughout professional healthcare education and embraced by the students. Clinicians should strive for excellence utilising a variety of continuing professional development opportunities. However, achieving high-level concussion-related self-efficacy might not be possible without joint support of stakeholders and mentors within clinicians’ professional education and work environments