3789 research outputs found
Sort by
Impact of hardware impairments on the physical layer security of cell-free massive MIMO
Get PDFThe development of new technologies and applications such as virtual reality, ultra-highdefinition video conferencing, and Internet of Things (IoT) has caused a substantial increase
in the demand for higher data rate in cellular systems. Massive Multiple-Input MultipleOutput (MaMIMO) is a reliable solution to fulfill this demand, not only providing higher data
rates, but also offering enhanced coverage and network capacity. These aspects are essential
to accommodate the rapidly increasing number of mobile subscribers with each passing
year. However, the swift progression of wireless communication technologies, including
fifth-generation (5G) networks and beyond raises a critical concern: ensuring the security of
these systems.
This thesis focuses on enhancing the security of Cell-Free Massive Multiple-Input
Multiple-Output (CF-MaMIMO), an advanced extension of MaMIMO. It uses a Physical Layer Security (PLS) technique which involves beamforming artificial noise (AN) in the
null of the users. Previous studies have demonstrated that implementing PLS techniques
always enhance the security performance of wireless communication systems. However,
these studies often overlook a crucial aspect: the impact of hardware impairments (HWIs).
They assume ideal transceivers in their research, neglecting the practical implications where
hardware non-idealities can significantly impact system security. Therefore, this thesis
analyzes the impact of HWIs on security performance based on broadcasting AN as a PLS
technique in CF-MaMIMO systems For this purpose, the Signal-to-interference-plus-noise
ratio (SINR) of the legitimate users and Signal-to-noise ratio (SNR) of the eavesdroppers
is derived considering HWIs in the implementation of AN broadcasting. Contrary to existing literature, it is demonstrated in this thesis that in certain instances, the AN leads to degradation in the security performance of the system due to HWIs. The findings of this
study reveal that fluctuations in the hardware quality of users, eavesdroppers and access
points (APs) directly affect the system’s security. Furthermore, these findings emphasize the
significance of considering hardware quality when applying PLS techniques by broadcasting
AN to maximize security performance
Towards accessible healthcare: machine learning-enabled diagnosis of Alzheimer’s disease
Get PDFAlzheimer’s disease poses a critical challenge to public health with an increasing prevalence among the aging population worldwide. The research question is whether machine
learning-based solutions could be a reliable, cost-effective, and non-invasive alternative to
existing biomarker tests. This thesis presents two machine learning-based approaches to
diagnosing Alzheimer’s disease using Magnetic Resonance Imaging (MRI) and blood-based
biomarkers. The first approach aims to train machine learning models on volumes of brain
regions from MRI to classify patients into three classes: Alzheimer’s Dementia (AD), Mild
Cognitive impairment (MCI) and Normal Control (NL). Pretrained weights of a well-known
CNN-based brain segmentation model were used in segmenting the hippocampal, parahippocampal, ventricles, entorhinal and cerebral white matter from MRI of patients, and their
volumes were estimated. The volumes and demographic data of the patients were subsequently trained on SVM and KNN models, and their performance was recorded.
The second approach aims to design efficient feature selection methods to identify relevant feature panels to identify individuals in the early stages of Alzheimer’s accurately. Two
feature selection methods were introduced. The first method ranks features according to
their dependence on the diagnosis, determined using metrics such as Mutual Information,
Symmetric Uncertainty and Cramer’s V. Panels are formed in this method by iteratively
selecting the top features and increasing the panel size. The second method filters out irrelevant features using the Euclidean distance between the class means of each feature and
applying a threshold. [...
Dialogue in enhancing EFL writing proficiency through collaborative synchronous and asynchronous computer-mediated communication
Get PDFWriting is considered to be one of the most challenging skills for English as a Foreign
Language (EFL) learners in China. To improve students’ writing skills, this study explored a
pedagogical strategy that blends collaborative writing with the use of computer-mediated
communication (CMC) technology in EFL contexts. Despite solid evidence supporting the
benefits of computer-mediated collaborative writing (CMCW) tasks in L2 writing, little is known
about the influence of CMC modalities on the efficacy of CMCW tasks. This qualitative case
study, conducted within a sociocultural framework, particularly Swain’s concept of collaborative
dialogue, examined how different task modalities influence the effectiveness of CMCW tasks in
the Chinese EFL context. Sixteen EFL learners completed an online collaborative writing project
on Tencent Docs™ via two modalities: A synchronous CMC (SCMC) modality entailing text
chat in WeChat™ and an asynchronous CMC (ACMC) modality with delayed interactions on
Tencent Docs™. Following the project, semi-structured interviews involving stimulated recall
were conducted to investigate the learners’ perceptions regarding these two patterns of
communication. The analysis of these interviews, coupled with insights gleaned from the
learners’ reflective journals collected throughout the writing project, indicated that although both
SCMC and ACMC modalities were perceived beneficial for writing development, EFL learners
had more positive learning experiences in the SCMC modality than in the ACMC one. The
findings revealed several challenges that require attention when implementing CMCW in EFL
teaching contexts, particularly where students possess weaker English proficiency and little
collaborative writing experience. EFL instructors are suggested to provide more training sessions
and offer appropriate guidance and feedback throughout the collaborative writing process
Describing the similarities and differences in songbird communities between harvested and wildfire-origin stands in Northwestern Ontario, Canada
Get PDFIn Ontario, sustainable forest management is mandated by the CFSA. Natural
disturbance emulation is viewed as method that improves sustainability in managed
forests. However, few studies have attempted to measure the effectiveness of natural
disturbance emulation with respect to maintaining ecological integrity. Using songbird
data that was collected in 2021 through the deployment of 96 Acoustic Recording
Devices on 157 sample plots in the Dog River-Matawin Forest Management Unit, an
analysis was conducted to investigate the similarities and differences in song
communities between wildfire-origin (n = 90) and harvest-origin (n = 67) stands.
Community-level indices (richness, abundance, and Shannon’s diversity index) were
calculated for multiple age classes representing different stand development stages for
five different forest species compositional groupings. It was found that in natural stands,
regardless of forest type, there was an increase in bird species richness, abundance,
and diversity as the forest matured. Managed stands supported a similar richness,
abundance, and diversity as natural stands. Where compared, natural and managed
stands had different community assemblages. Downy woodpecker (Dryobates
pubescens) was entirely absent from management stands but was present in natural
stands, suggesting that there may be functional differences between the two origin
types. Managed stands may have a lower density of standing deadwood with specific
dimensions preferable to the downy woodpecker. The retention of size-specific standing
deadwood during harvesting may benefit the downy woodpecker in managed forests
Insights into subtype selectivity of aurora kinase ligands from molecular dynamics simulation
Get PDFAurora kinases are phosphotransferase enzymes that play essential roles in cell division. There
are three members of Aurora kinases in mammalian cells: Aurora A, Aurora B and Aurora C. The
overexpression of Aurora kinases in diverse cancer cells make them promising targets in cancer
therapy. Aurora kinases show highly conserved homology, having four different residues in the
active site: Leu215, Thr217, Val218, and Arg220 in Aurora A (Arg159, Glu161, Leu162 and
Lys164 in Aurora B). Therefore, understanding Aurora kinase inhibitor selectivity remains a top
priority for kinase inhibitor design.
The utilization of molecular dynamics simulations for kinase selectivity studies could provide
insights into ligand-protein interactions, including key residues, predominant free energy
contributions, and interaction types, facilitating the design of subtype-selective inhibitors. To
elucidate the subtype selectivity mechanism of Aurora kinase A and B, molecular docking was
employed to construct complex structures. Subsequent MD simulations were conducted for
complexes of Aurora A and B with selective inhibitors LY3295668, MK-5108, and Alisertib, as
well as Aurora B selective inhibitor GSK-1070916 and pan-inhibitor Danusertib. The analysis
included RMSD, average structure determination, MM/PBSA-derived binding free energy, and
decomposition analysis, elucidating favorable or unfavorable residue contributions within the
active site. For Aurora A selective inhibitors (LY3295668, MK-5108, and Alisertib), the residue
Thr217 and Arg220/137 emerged as crucial for selectivity, with the carboxylate group being the
predominant functional group contributing significantly to binding free energy in these
compounds. Conversely, GSK-1070916's selectivity for Aurora B was attributed to Arg159 and
Asp218, with its tertiary amine with methyl group being key functional groups. These findings
on subtype selectivity mechanisms hold promise for the development of highly selective Aurora
kinase inhibitors, offering a less toxic anti-cancer strategy
A review of the Toronto Zoo's head-starting program for recovery of the Blanding's turtle in Rouge Valley National Urban Park
Get PDFConservation reintroduction programs are valuable tools in supporting endangered or
extinct species in the wild. With the many ways humans are causing adverse environmental
impacts, it is crucial that we put effort into reversing our adverse effects to avoid large-scale
irreversible changes to ecosystems. Places like zoos and sanctuaries already have facilities and
staff extensively trained in caring for animals. These locations can be the key institutions to
support various wildlife conservation projects. The Blanding’s turtle head-starting program at the
Toronto Zoo and the turtle reintroduction into Rouge Valley National Urban Park are successful
steps in restoring a population of an endangered species. The year that the individuals were
released over the period 2014-2020 did impact the turtles’ chances of survival, with particularly
low survival in 2020, but there was equal success with male and female releases and variable but
equal success with hard and soft releases
Exploring name-based bug detection in Python
Get PDFNames of source code elements provide useful contextual information about the code and
development tasks. Prior studies leverage the similarity between the names of arguments
and method parameters to detect bugs that are caused by accidentally swapping arguments
while calling methods. This requires establishing the mapping between method calls and
their definitions. However, it is a challenging task to establish the mapping because of the
complexity involved with the process (e.g., missing external libraries). This thesis aims to
understand the performance of name-based argument-related bug detection techniques in
Python, a popular general-purpose, statically typed programming language.
Towards this direction, this thesis conducts a study that first investigates the similarity
between arguments and their method parameters in Python code. The above step follows by establishing the mapping of method calls to their definitions and evaluating the
performance of existing name-based techniques to detect swapping argument-related bugs
in Python. Finally, a technique has been developed that uses argument usage patterns
and expression types in source code with name-based similarity matching to improve the
performance of detecting argument-related bugs. Evaluation of the proposed technique
with a large collection of open-source Python projects shows that the technique can detect
argument-related bugs with high accuracy even when the method definitions are missing.
One potential solution to prevent argument-related bugs from occurring is to use code completion. An argument recommendation system suggests method arguments as a developer
types the code. Thus, the second part of the thesis focuses on completing arguments of
method calls. In particular, this thesis investigates the efficacy of large language models in
recommending arguments for API (Application Programming Interface) method calls
Fracture behavior of natural fiber-reinforced cemented paste backfill under mode-I, mode-II, and mode-III loading: Effect of fiber content and fiber length
Get PDFThe mechanical stability of mine backfill materials is crucial for the safety of mining
personnel and production efficiency. When placed into mined-out voids, known as stopes,
mine backfill materials are required to provide reliable secondary ground support, which
is subjected to finite deformation loading, especially in deep mines. However, due to the
quasi-brittle characteristics of cemented paste backfill (CPB), these materials possess
very limited post-peak resistance. Enhancing the post-peak engineering performance of
CPB is achievable through natural fiber reinforcement techniques. In this study, hemp
fibers were selected for their abundant availability in Canada. To investigate their
effectiveness in terms of fiber reinforcement, four different fiber lengths (5mm, 10mm,
20mm, and 30mm) and four fiber contents (0.25wt%, 0.5wt%, 1wt%, and 1.5wt%) were
employed to prepare the natural fiber-reinforced CPB (NFR-CPB). A series of mechanical
tests, including semicircular bend (SCB) tests and end-notched disc bend (ENDB) tests,
along with scanning electron microscope (SEM) observations, were conducted on NFRCPB and control CPB (without fiber reinforcement) at 7 days, 28 days, and 90 days. The
results revealed that hemp fiber reinforcement can influence pre-peak behavior and
effectively enhance post-peak resistance. Additionally, the results showed that increasing
the hemp fiber content and length improved the fracture energy, ductility, and fracture
toughness of NFR-CPB. Therefore, the proposed hemp fiber reinforcement approach can
be considered a promising method for CPB technology in deep mining applications
Fracture behavior and properties of fiber-reinforced cemented paste backfill under different displacement rates
Get PDFCemented paste backfills (CPB, comprising waste tailings, hydraulic binder, and water)
are essential for safe and efficient underground mining production. Once delivered into
underground excavations, CPB undergoes complex field conditions. Maintaining the
engineering performance and mechanical stability of mine backfill mass, including fibers,
can significantly enhance its mechanical behaviors (especially post-peak resistance) and
CPB properties. Nevertheless, the field loading conditions are commonly featured in
various loading rates. Therefore, to ensure backfill mass stability, it is essential to fully
understand the effect of displacement rates on the mechanical behaviors and properties
of fiber-reinforced-CPB (FR-CPB). This study examined the effect of different
displacement rates (0.2mm/min, 1mm/min, 5mm/min, and 10mm/min) on the fracture
behavior and properties of FR-CPB, which were prepared with three polypropylene fiber
lengths (6mm, 13mm, and 19mm), and four fiber content (0.25%, 0.5%, 1%, and 1.5%)
at three different curing times (7 days, 28 days, and 90 days). A comprehensive testing
program was designed and performed, including semi-circular bend tests, end-notched
disc bend tests, and SEM observations. The results demonstrate that load-displacement
curves are sensitive to the changes in the displacement rate, with a higher displacement
rate resulting in a higher peak load and post-peak resistance load. Moreover, it was also
found that, compared with the influence of fiber content, the fiber length variation can
interfere with the displacement-rate dependency of fracture properties to a greater extent.
Furthermore, it is interesting to observe that excessive usage of fibers can cause material
stiffness, fracture toughness, and fracture energy degradation. loading rate (1mm/min) has been identified, which can cause significant changes in the
rate of change in fracture properties. In terms of optimum usage of fibers, the results show
that a fiber length of 13mm and a fiber content of 0.5% can maximize the fiber
reinforcement effect on the improvement of fracture properties of CPB under different
displacement rates. Therefore, the findings from this study can potentially promote the
successful implementation of fiber reinforcement techniques in the mine backfill
operation
Characterization of alteration and mineralization of the Moss gold deposit, Shebandowan greenstone belt, Northwestern Ontario
Get PDFThe Moss Au deposit is an orogenic-style gold deposit hosted in felsic to intermediate rocks of
the western Shebandowan Greenstone Belt, close to the terrane boundary between the
Wawa-Abitibi terrane and the Quetico metasedimentary basin, ~120 km west of Thunder Bay.
The deposit has an inferred mineral estimate of 140.07 Mt of ore averaging 1.09 g/t Au, which
yields 4.91 Moz (Goldshore, 2024). The majority of the gold is hosted within diorite and dacite
and is localized by shear zones and an array of quartz-carbonate-pyrite veins. The central-felsic
metavolcanic belt of the Shebandowan Greenstone belt comprises felsic to intermediate units
surrounded by late granitic intrusions, such as the Burchell Lake and Moss Lake stocks. This
study focused on characterizing the alteration and mineralization at the Moss deposit and
investigating any correlation between alteration and gold mineralization. A combination of
petrography, geochronology, geochemistry, and mineral chemistry was used to achieve the
objectives of this study.
Alteration occurs in different styles and intensities but generally comprises albite, biotite,
sericite, chlorite, carbonate, and epidote alteration. Sulfide minerals are dominated by pyrite
with minor chalcopyrite, sphalerite and molybdenite. Sulfide abundance is commonly 2 – 10%
of the samples but can be up to 15% within sulfide-rich veins. Disseminated and vein-hosted
pyrite are the two main textures in which pyrite occurs within the host rocks. A total of 12 vein
types were observed, with quartz and carbonate being the most dominant veins occurring
together in five of the vein types. Using the observed textural and crosscutting relationships of
the alteration, sulfides, and veins, a paragenetic sequence was developed, highlighting the
secondary processes associated with the formation of the Moss Lake deposit. Deformation
textures were observed in early and late alteration phases, suggesting a long deformation
history that was broadly coeval with mineralization.
Quartz-carbonate-pyrite ± sericite ± chlorite ± epidote veins are host to most of the observed
gold occurrences, and are common within or in proximity to shear zones. Gold was rarely
associated with disseminated pyrite away from veins. [...