21090 research outputs found
Sort by
Design and Development of Instrumented Foot Form for Testing of Metatarsal Protective Footwear
No full textProtective footwear with metatarsal guards can play a critical role in preventing high energy impact injuries to the foot. Currently, there is a wide variety of different metatarsal guard types, each of which must pass impact testing standards before being used. These current impact testing standards to assess the efficacy of protective footwear, such as ASTM F2412, often use deformation of an internal material like wax as the performance indicator for metatarsal guards. However, previous studies have shown that deformation is not a great predictor of metatarsal fracture injury risk. Therefore, the overall goal of this work was to develop and validate a biofidelic instrumented foot surrogate for evaluating the impact performance of metatarsal protective equipment. A combination of computational and experimental methods were first used to identify a surrogate material that demonstrates a similar force-deformation response to the human foot. Next, design for additive manufacturing parameters, such as lattice structures, infill density, and unit cell size, were used to produce a foot form with a biofidelic impact response. This design process used an
iterative methodology to develop three prototypes of additively manufactured surrogates using engineered hyperelastic material and embedded load cells. A series of drop tower tests were conducted using ASTM methodology, and the force and deformation for each prototype was compared to cadaveric data reported in the literature. Three types of metatarsal guards were drop tested with the prototypes, and transmitted forces were recorded through an embedded force measurement system. Results from impact testing showed that all developed prototypes provided a closer match to cadaveric force-deformation behaviour, with Prototypes I and III performing slightly better. Metatarsal guard performance results were limited as the load-sensing equipment was overloaded. Prototype I’s load-sensing method was unreliable. Prototype II reported the worst performance for soft metatarsal guard footwear. Prototype III demonstrated that soft and hard metatarsal guards offer better protection than boots without guards. In conclusion, this study presented a novel foot surrogate for metatarsal impact resistance testing. Further studies are required to refine the design to improve the force measurement system and ensure a better fit of the foot surrogate inside protective footwear
Resource Allocation and Task Scheduling for Integrated Sensing and Communications
No full textIntegrated Sensing and Communications (ISAC) has emerged as a promising paradigm for future Sixth-Generation (6G) wireless networks. In this paradigm, wireless networks can have both Sensing and Communication (SAC) capabilities using shared network resources. ISAC not only enables the provisioning of SAC services but also has the potential to enhance their performance: end-user devices can offload sensing data collection tasks to Access Points (APs) co-located with edge servers and upload raw or preprocessed sensing data to powerful edge servers for high-performance processing; meanwhile, APs can leverage contextual information about communication tasks obtained through sensing, such as the visibility of virtual content in mobile Augmented Reality (AR) streaming, to enhance communication efficiency. The interesting issue is to efficiently utilize network resources to optimize SAC service performance in the presence of high and spatiotemporally varying service demands. However, the main technical challenges are: 1) how network resource allocation and SAC task scheduling are proactively determined to enable efficient coordination between APs and mobile end-user devices for achieving satisfactory service performance; 2) how an end-user device adaptively offloads computation-intensive Deep Neural Network (DNN)-based sensing tasks to an edge server to optimize task performance, under dynamic task arrival, task processing, and server workload statuses; and 3) how an AP efficiently acquires contextual information about communication tasks through sensing individual mobile AR users and dynamic environment for resource-efficient mobile AR streaming.
In this thesis, we develop efficient resource management schemes for ISAC, including resource allocation and task scheduling, to address the above three technical challenges. First, we investigate proactive resource management for ISAC, determining the reservation of radio and computing resources, the active probability of mobile devices for communications, and the partitioning of sensing regions. To cope with the non-stationary spatial distributions of mobile devices and sensing targets, which can result in the drift in modeling the distributions and ineffective resource management decisions, we construct Digital Twins (DTs) of the network slices for individual SAC services. In each DT, a drift-adaptive DNN-empowered statistical model and an emulation function are developed for the spatial distributions in the corresponding slice, which facilitates closed-form decision-making and efficient validation of a resource management decision, respectively. Numerical results demonstrate that the proposed scheme can significantly enhance service satisfaction ratios and reduce resource consumption compared to benchmark schemes. Second, we investigate task offloading for DNN-based sensing data processing. Particularly, we consider that an end-user device stochastically generates and adaptively offloads DNN-based sensing data processing tasks to an AP co-located with an edge server. To adapt to the dynamic on-device and edge server workload status, leveraging the multi-layer and multi-exit architecture of the considered DNNs, an offloading decision for each sensing task is made on whether and when to stop on-device task processing and offload the task to the edge server to complete the processing. Two DTs are constructed to evaluate all potential offloading decisions for each sensing task, which provides augmented training data for a machine learning-assisted decision-making algorithm, and to estimate the task processing status at the device, which avoids frequently fetching the status information from the device and thus reduces the signaling overhead. Simulation results demonstrate the outstanding performance of the proposed task offloading scheme in terms of balancing sensing result accuracy, delay, and energy consumption. Third, we propose an efficient resource allocation scheme in sensing-assisted mobile AR streaming. In specific, we consider that the position and surrounding environment of an AR user can be captured via sensing to extract contextual information for AR streaming, i.e., the visibility of virtual content. The goal is to minimize the overall radio resource consumption for delivering virtual content visible to the AR user by properly determining the radio resource allocation for user positioning and environment mapping. To this end, we first develop a mathematical model to estimate the content visibility uncertainty and the content delivery resource consumption. We then generate a reference resource allocation decision that guides a deep reinforcement learning-based decision process to efficiently adapt to non-stationary user and environment dynamics. Trace-driven simulations demonstrate that the proposed scheme significantly reduces radio resource consumption for delivering virtual content visible to an individual AR user, compared to benchmark schemes.
In summary, we have proposed efficient resource management schemes for ISAC that optimize SAC service performance with efficient resource utilization and practical operational complexity. The research results from the thesis provide valuable insights into the design of scalable and adaptive ISAC systems that seamlessly unify sensing, communications, and intelligence in the future 6G
State estimation using machine learning
No full textState estimation refers to determining the states of a dynamical system that evolves under disturbances, based on noisy measurements, partially known or unknown initial condition, and a known system model. JRNs have a structure that mimics that of a dynamical system and are thus attractive for estimator design. We show that a JRN performs better than an EKF and UKF for several examples. We also provide an input-to-state stability analysis of the error dynamics of JRNs. The stability of the error dynamics of several examples is shown.
We then extend the Jordan structure to long-short-term memory networks to obtain a JLSTM which, as we show in several examples, is comparatively more robust to changes in initial conditions and noise and performs better than a EKF and PF. It also trains faster than an ELSTM for state estimation when trained to achieve a similar normalized MSE.
We also compare a shallow and deep JLSTM and observe that they perform almost similarly in terms of average error across time-steps and MSE but the deep JLSTM takes longer to train due to more layers.
We also train a JLSTM with a modified maximum likelihood equivalent loss function(JLSTM-ML). We observe that for Gaussian initial conditions and disturbances, the average error at each time step is best for estimates of JLSTM-ML. It is also the most robust to changes in initial conditions and disturbances in the systems considered. The measures, time taken to train, time taken to test, mean squared error, and average error at each time-step were used for comparison for various networks.
We discretized the following systems to use as examples in data generation, training, and testing: mass-spring system, down pendulum, reversed Van der Pol oscillator, Galerkin approximation of Burger's partial differential equation and Kuramoto-Sivashinsky partial differential equation
Modeling and optimal operation of sustainable thermoelectric microgrids with phase-change material thermal system
No full textThe final publication is available at Elsevier via https://doi.org/10.1016/j.segan.2025.101814. © 2025. This manuscript version is made available under the CC-BY-NC-ND 4.0 license http://creativecommons.org/licenses/by-nc-nd/4.0/This paper proposes an Energy Management System for a thermoelectric microgrid that incorporates the modeling of a unique Phase-Change Material-based thermal system, capable of operating in both active and passive modes to minimize operating costs while guaranteeing thermal comfort, while properly considering the microgrid’s thermal power requirements and indoor temperature control. The proposed model also includes a detailed thermal representation of buildings to consider relevant thermal sources and room heat exchange, as well as heat pumps, water tanks for thermal storage, and battery degradation. A Model Predictive Control approach is used to address uncertainties in demand and environmental conditions. The proposed Energy Management System model is applied to the Energy Smart Home Lab microgrid located at the Karlsruhe Institute of Technology, in Germany, taking into account the specific characteristics of the microgrid’s components, expected energy consumption, and indoor temperature control requirements. Simulation results demonstrate the feasible application of the developed Energy Management System for the optimal operation of the actual microgrid considered, illustrating the thermoelectric microgrid’s power balance and temperature fluctuations of the associated components, with particular emphasis on the operation of the Phase-Change Material system, to showcase its active and passive thermal contribution under extreme weather conditions.This work has been supported by the Natural Sciences and Engineering Research Council of Canada (NSERC)
Growing Beyond Fiction: Solarpunk as an Ethos of Care
No full textThis thesis explores the intersection of care theory, a branch of feminist ethics that emphasizes relationality, responsiveness, and the moral significance of care, with the emerging aesthetic and political movement of solarpunk. Care theory challenges traditional ethical frameworks by focusing on interdependence, context, and the transformative role of emotions. Meanwhile, solarpunk, originating as a science fiction subgenre and expanding into a broader social imaginary, envisions sustainable futures grounded in environmental stewardship, technological innovation, and post-capitalist forms of cooperation. I argue that solarpunk not only embodies the central themes of care ethics but also extends them by offering a practical, imaginative, and mobilizing ethos for contemporary political and environmental challenges. Through an interdisciplinary methodology, this project demonstrates how solarpunk’s narratives and practices can respond to critiques of care ethics. Finally, by analyzing the figure of the cyborg as developed in feminist theory and reinterpreted through solarpunk, I contend that the cyborg operates as a metaphorical caregiver—dissolving boundaries between human, technology, and environment—thereby expanding the ontology of care itself. In doing so, this thesis positions solarpunk as a vital and actionable framework for imagining and enacting a livable, care-centered world
Cuts in Optimization and Approximation: Generalized MIR Cutting Plane and Integrality Gap Bounds for the Planar Multicut Problem
No full textThis thesis has two main themes, both centered around the role of cuts in integer programming and approximation algorithms.
In the first part, we investigate the complexity of split closures in mixed-integer sets in~ with two integral variables and one continuous variable. While it is known that in~ the split closure admits a polynomial-size description, extending this result to the mixed-integer setting in~ poses new challenges. For a rational polyhedron~ and integer index set~, we denote by~ the mixed-integer hull of with respect to , and by~ the intersection of all split cuts with respect to . We make progress toward understanding the structure of~ by proving the existence of a polyhedron~ such that~, where~ can be described by a polynomial number of inequalities relative to the input size of~. In addition, we introduce a generalized mixed-integer rounding (MIR) procedure that begins with systems of two equations, rather than a single equation as in the classical setting. This leads to a new family of valid inequalities for mixed-integer sets, whose derivation relies on structural results for split closures in~.
The second part focuses on approximation algorithms for the minimum multicut problem in planar graphs and certain subclasses thereof. We analyze the integrality gap of the natural LP relaxation within the framework of small-diameter decompositions. By introducing novel tools, we improve the best known lower bound on this integrality gap from to for the family of cactus graphs, which in turn yields an improved lower bound of the integrality gap for planar graphs. Complementing this, we establish an upper bound of for cactus graphs, improving the previous bound of . We further obtain tight bounds in special subclasses: the integrality gap is exactly for unicyclic graphs and for path cactus graphs. Extending the study to outerplanar graphs, we prove an upper bound of on the integrality gap by designing a randomized rounding algorithm that transforms the optimal fractional solution of the LP relaxation into a feasible multicut whose expected cost is at most times the cost of the optimal fractional solution. Together, these results sharpen our understanding of the minimum multicut problem and its LP relaxation in planar graphs
Additivity of the Quantum and Classical Capacities of Quantum Channels
No full textQuantum channels enable communication through the transmission of quantum states. Quantum Shannon theory investigates these channels, aiming to characterize their capacity for information transmission under various conditions. While this characterization is well-established for classical communication channels, quantum channels exhibit significantly more complex and mathematically intricate behavior, making a complete understanding elusive. A key challenge is the phenomenon of non-additivity, where combining quantum channels can enhance information flow by leveraging quantum effects. In this work, we focus on two types of non-additivity: those of classical capacity and quantum capacity.
We present new constructive counterexamples demonstrating the non-additivity of the minimum output p-Renyi entropy for p>2. These examples achieve non-additivity at lower values of p than previously known constructions of the same dimension. We also show that several plausible generalizations of antisymmetric spaces -- such as through alternative symmetries or higher tensor powers -- cannot produce non-additivity using current techniques.
Additionally, we advance the study of resonant multilevel amplitude damping channels. We analytically derive their degradability regions, previously inferred using a heuristic assumption supported by numerical evidence, and formulate conjectures on their capacity based on our own numerical evidence. Specifically, we conjecture that their coherent information is optimized on diagonal states and that they are always weakly additive. However, we find that coherent information activation is possible, as strong non-additivity arises in certain regions when combined with erasure channels
Designing Porous Polymer Systems for Water Treatment Applications
No full textIncreasing pollution and contamination of the World’s water bodies come with great concern over potable water safety and accessibility. Current solutions for water treatment often have large carbon footprints or are too expensive to scale up effectively. These shortcomings warrant the exploration of new and effective methods of water treatment. Polymer-based solutions offer lightweight, scalable, and inexpensive methods for water filtration while being minimally intrusive to the surrounding environment. In particular, porous polymeric materials have garnered considerable attention due to their high specific surface area, which enables them to have enhanced interactions with their target analyte.
This thesis presents two such types of porous materials: nonwoven fabrics and three-dimensional (3D) printed filters. The first section of this thesis focuses on nonwovens, a type of fabric comprised of bonded, interlocking, randomly oriented fibers. Nonwovens can be used as topically placed sorptive mats for the removal of pollutants, or as a pass-through filter for the separation of water from the pollutants. Here, the unique oil gelation properties styrene-ethylene-butylene-styrene (SEBS) block copolymer are leveraged for the creation of melt-blown nonwovens for oil-water separation applications. The poor processability of SEBS, due to its elastomeric nature, was overcome through highly optimized processing parameters to create fine diameter, highly porous nonwoven mats. These mats possessed exceptional lipophilicity and oil-water separation properties due to the oil-soluble midblocks of SEBS that created a semi-solid gel capable of retaining all oil it came into contact with.
The latter section of this thesis focuses on 3D printing, specifically fused deposition modelling (FDM), for the creation of flow-through filters for microplastic capture. 3D-printed parts are often very smooth, greatly limiting their surface area and ability for microplastics to become lodged on their surface. To overcome this, a sacrificial additive was added to the base polymer matrix that could be etched out, creating a highly porous surface that greatly improved the filtration efficiency of the printed filters. Pressure-sensitive adhesives (PSAs) were also explored and were found to further bolster the filtration capabilities of the filters. This is due to the added tack and non-covalent interactions that more strongly hold microplastics to the surface of the filters. The findings from these studies demonstrate a promising direction for utilizing porous polymer systems in water treatment applications
An Optimal Knowledge Retention Framework for Continual Learning in Data Stream Scenarios
No full textIn the field of time series and data stream analysis, neural networks (NNs) have demonstrated excellent performance in predicting current and future states of dynamic systems. However, forgetting a previously learned information by NN when training the model on new data can be a significant challenge in having a reliable prediction, a problem that is known as catastrophic forgetting (CF) in neural networks. Unfortunately, retraining the model with both historical and new data is often impractical due to computational complexity and storage constraints, particularly in large-scale applications. One of the most prominent examples is automotive systems, where dynamic environments, such as changing road conditions or driving scenarios, require continuously updating the existing information based on new data.
The main objective of this thesis is to propose a continual learning method that can efficiently train a neural network model on newly collected data while preserving previously acquired information. A novel framework based on memory-based continual learning approaches is developed, consisting of two critical tasks: optimal sampling of the old data to store in memory, and optimization. First, the proposed method aims to identify the most relevant and informative memories for old dataset, which are then contributed in future learning to preserve the previously learned information. The proposed method is developed in both univariate and multivariate time series prediction scenarios. Second, a proper optimization technique is used in each training epoch to minimize the loss function by modifying the network parameters, ensuring that NN is capable of successfully integrating new input while maintaining historical information.
Additionally, a hybrid state estimation framework is introduced, leveraging the selected memory points to detect distribution shifts in real-time within the incoming data stream. When the estimator detects unfamiliar patterns that may degrade the predictive performance of the neural network, it adaptively switches to a model-based estimator to ensure robust and reliable estimation under the newly encountered conditions. A variety of neural network models and architectures are explored and compared to provide a comprehensive analysis and to evaluate their effectiveness in state estimation tasks. Furthermore, uncertainty analysis is conducted using conformal prediction, enabling quantification of the neural network’s predictive uncertainty after training on each task and comparison to a conventional batch learning baseline.
The proposed framework is applied to both univariate and multivariate scenarios for estimating vehicle longitudinal and lateral velocities, incorporating new driving maneuvers into the previously trained neural network model. Experimental datasets comprise of sensor measurements from an electric Equinox vehicle. The effectiveness of the method is evaluated by examining the performance of the model in training on new information as well as the impact of forgetting on previously acquired knowledge as new tasks are incrementally introduced. The findings of this study suggest that the developed continual learning framework is capable of efficiently training the model on new data while preserving the prediction accuracy on previous data. The time efficiency of the proposed method is an important advantage, as it enables the neural network to adapt to new tasks quickly without a significant computational overhead
Exploiting Zero-Entropy Data for Efficient Deduplication
No full textAs the volume of digital data continues to grow rapidly, efficient data reduction techniques, such as deduplication, are essential for managing storage and bandwidth. A key step in deduplication is file chunking, which is typically performed using Content-Defined Chunking (CDC) algorithms. While these algorithms have been studied under random data, their performance in the presence of zero-entropy data, where long sequences of identical bytes appear, has not been explored. Such zero-entropy data are common in real-world datasets and introduce challenges for CDC in deduplication systems.
This thesis studies the impact of zero-entropy data on the performance of both hash-based and hashless state-of-the-art CDC algorithms. The results show that existing algorithms, particularly hash-based ones, are inefficient at detecting and handling zero-entropy blocks, especially when these blocks are small, which reduces space savings. To address this issue, I propose ZERO (Zero-Entropy Region Optimization), a system that can be integrated into the deduplication pipeline. ZERO identifies and extracts zero-entropy blocks prior to chunking, compresses them using Run-Length Encoding (RLE), and stores their metadata for later reconstruction. ZERO improves deduplication space savings by up to 29% without impacting throughput