UWSpace

UWSpace is the University of Waterloo’s institutional repository for the free, secure, and long-term home of research produced by faculty, students, and staff.

Depositing Theses/Dissertations or Research to UWSpace

Are you a Graduate Student depositing your thesis to UWSpace? See our Thesis Deposit Help and UWSpace Thesis FAQ pages to learn more.

Are you a Faculty or Staff member depositing research to UWSpace? See our Waterloo Research Deposit Help and Self-Archiving pages to learn more.

Photo by Waterloo staff

Communities in UWSpace

Select a community to browse its collections.

Recent Submissions

  • Thumbnail Image
    Item type: Item ,
    Load Variation Resilient and Average Efficiency Enhanced Power Amplifiers for 5G/6G Beamforming System
    (University of Waterloo, 2026-05-06) Yu, Hang
    The deployment of Fifth Generation (5G) and Sixth Generation (6G) infrastructure relies heavily on high-frequency beamforming architectures to deliver high data rates and spectral efficiency. However, the physical realization of these systems faces critical challenges: the need for high circuit integration, energy efficiency under high Peak-to-Average Power Ratio (PAPR) signals, and robustness against dynamic load variations inherent in large-scale arrays. This doctoral thesis addresses these requirements through the development of three advanced integrated circuit objectives, progressing from theoretical derivations in load variation resiliency (Voltage-Standing-Wave-Ratio (VSWR) resiliency) to front-end architectural synthesis for Power Amplifiers (PAs). To improve circuit integration and performance for Time-Division Duplex (TDD) operation in beamforming systems, Chapter 3 focuses on the co-design of a Transmit/Receive (T/R) Front-End Module (FEM). Traditional FEMs suffer from insertion loss and area overhead due to additional Single-Pole Double-Throw (SPDT) switches. To resolve this, this work presents an architecture that integrates a Doherty Power Amplifier (DPA) in the transmit path, which also functions as a switchless T/R isolation network during receive operation. On the receiver side, an embedded switching network maximizes isolation and bandwidth while jointly optimizing the overall FEM performance and integration trade-offs. A 39 GHz prototype was fabricated using the GlobalFoundries 45nm Silicon-On-Insulator (SOI) CMOS process and achieves a Transmit (TX) mode gain of 15 dB, a saturated output power of 20 dBm, and a Power-Added Efficiency (PAE) of 23%/15% at peak and 6-dB back-off, respectively. In Receive (RX) mode, it delivers 20 dB of gain, a 4.5-dB noise figure, and an input 1-dB compression power of -16.5 dBm while consuming 32 mW. Occupying a core area of just 0.5 x 0.75 mm^2, this architecture demonstrates a highly competitive efficiency-noise-integration trade-off, achieving state-of-the-art performance for high-frequency FEMs. While integration and performance improvements are critical for beamforming systems, the load variation induced by antenna mutual coupling in Large-Scale Antenna Arrays (LSAAs) presents another critical challenge. Chapter 4 proposes a dual-mode PA design, utilizing different gate biasing to reconfigure its operational state. The work features a 'VSWR resiliency mode' to maintain robust performance under high load mismatch, and an 'Output-Back-Off (OBO) efficiency enhancement mode' to maximize efficiency under minimal load variation. Central to this design is a novel combiner network synthesized to support two distinct operational regimes: it can emulate the characteristics of balanced architectures (symmetric drain currents) for load-variation resiliency, or Doherty load modulation (asymmetric drain currents) for efficient OBO operation. A 29 GHz prototype was fabricated using GlobalFoundries' 22nm Fully-Depleted SOI CMOS process to validate the concept. Under a 50-ohm load, the VSWR-resilient mode achieves 16.5-dB gain, 12.5-dBm output power, and 18%/7.5% PAE at peak/6-dB OBO. The OBO efficiency-enhanced mode delivers 13-dBm output power with 19%/12.5% PAE at peak/6-dB OBO. Across a load of 2.5:1 VSWR over a 360-degree phase range, the VSWR-resilient mode exhibits only 0.5-dB average saturated-power degradation compared to 1~dB in the OBO efficiency-enhanced mode. Modulated measurements under varying VSWR loads further confirm the superior load-variation tolerance of the proposed architecture. The above-mentioned dual-mode approach offers flexibility between VSWR resiliency and efficiency improvements; however, communication protocols often demand simultaneous efficient and robust operation. Consequently, Chapter 5 unifies these requirements by establishing the theory and design of a "VSWR-Resilient DPA," extending the analytical framework of Chapter 4 to ensure robust performance across multiple power regimes. The analysis yields architectures that maintain the OBO efficiency profile of a DPA while simultaneously delivering load-variation insensitivity against Multiple-Input-Multiple-Output (MIMO) beamforming array mismatch. A prototype targeting 8 GHz is designed using a commercial MACOM GaN bare-die transistor on a multi-layer PCB substrate; however, due to procurement delays, experimental validation is deferred to future work. In EM circuit co-simulation, the architecture achieves 10-dB SS gain, 45-dBm saturated output power, and 49%/35% PAE at peak/6-dB OBO under a 50-ohm load, while maintaining less than 1.5-dBm saturated power variation and 1.55× normalized Class-B efficiency at 6-dB OBO across different antenna loads on the 3:1 VSWR circle.
  • Thumbnail Image
    Item type: Item ,
    Sustainable Strategies for Arctic Lifelines: Funding Decisions in the Face of Climate-Change Uncertainty
    (University of Waterloo, 2026-05-06) GHOLAMI, HAMED
    Remote Arctic communities rely on ephemeral winter roads for the affordable delivery of essential goods. As climate change destabilizes the physical foundation of these supply chains, policymakers face a complex stochastic allocation problem: how to optimally divide a constrained budget between supply-side infrastructure investments (to extend road duration) and demand-side consumer subsidies (to bolster household purchasing power). In this thesis, we develop a stylized stochastic optimization model to analyze this trade-off, formulating a capacity-budget gap parameter to capture the dual-bottlenecks of physical throughput versus financial liquidity. We prove that the optimal funding strategy, serving as a stochastic hedge, is strictly bounded between the supply-constrained and demand-constrained deterministic solutions. Through comparative statics, we uncover counter-intuitive operational tradeoffs, including a price-affordability tradeoff and a logistics efficiency tradeoff, where improvements in supply chain economics rationally trigger infrastructure divestment due to an underlying income saturation effect. Furthermore, we analyze the compound effect of climate change: a secular decline in mean winter duration combined with rising interannual volatility. We demonstrate that for the most vulnerable communities, this compound shock pushes the system into a dilution state where higher volatility counter-intuitively reduces the optimal investment level. Our findings suggest that as climate uncertainty accelerates and mean operational windows shrink, relying on winter road infrastructure becomes economically unsustainable, necessitating a strategic policy pivot toward direct income support and alternative logistics.
  • Thumbnail Image
    Item type: Item ,
    Feature Representation for Sea Ice Mapping
    (University of Waterloo, 2026-05-06) Noa Turnes, Javier
    Sea ice monitoring is essential for climate research, Arctic navigation, and operational decision-making. Synthetic aperture radar (SAR) imagery is the primary sensing modality used by national ice services because of its independence of atmospheric and lighting conditions, and sensitivity to surface structure. However, SAR-based sea ice classification remains challenging due to spatially non-stationary statistics caused by incidence angle effects, seasonal transitions, and strong within-class variability. These factors complicate feature extraction and limit the robustness and transferability of conventional deep learning models. This thesis investigates feature representation learning for sea ice classification in SAR imagery through both supervised and self-supervised paradigms. The first contribution introduces a supervised semantic segmentation framework that integrates convolutional neural networks (CNNs), transformers, and unsupervised region segmentation. The proposed Irregular Tokens on Transformers (ITT) architecture forms multi-scale, homogeneous tokens using Iterative Region Growing on Semantics (IRGS) and applies self-attention to capture long-range spatial dependencies. A multi-task training scheme combines pixel-level and region-level loss functions, encouraging region-consistent feature representations while preserving fine-grained boundaries. Experiments on multi-season RADARSAT-2 scenes demonstrate improved overall accuracy, sharper boundary delineation, and reduced predictive uncertainty compared to a CNN baseline. An expert audit conducted by the Canadian Ice Service further supports the operational relevance and stability of the approach across freeze-up and melt conditions. While supervised learning delivers strong performance when annotations are available, SAR labeling remains costly and domain specific. The second contribution explores self-supervised pre-training toward a SAR foundation model for sea ice classification. By leveraging masked representation learning and multi-task objectives, the proposed framework learns transferable representations from unlabeled SAR imagery. The study evaluates whether large-scale pre-training alone is sufficient to address domain shifts across sensors and seasons, or whether task-specific adaptations remain necessary. Results show that self-supervised pre-training substantially improves downstream performance and generalization, but optimal accuracy is achieved when combined with structured fine-tuning aligned with sea ice semantics. Overall, this thesis demonstrates that robust sea ice classification fundamentally depends on how feature representations are learned, and provides principled strategies for improving scalability, generalization, and operational viability in Arctic SAR applications.
  • Thumbnail Image
    Item type: Item ,
    Leveraging Interactive Human–AI Collaboration Methods to Enhance Key Stages of Programming Workflows
    (University of Waterloo, 2026-05-06) Liu, Xuye
    Beyond writing code, programmers routinely move through several complementary tasks as they develop, refine, and share their work. These workflows typically involve recurring stages: understanding and documenting code, checking correctness and debugging, improving efficiency and scalability, and sharing results with others. Each stage has its own challenges: documentation often becomes outdated or inconsistent with evolving code, debugging can be time-consuming and opaque, performance improvements require balancing competing goals (e.g., speed, memory, and clarity), and communicating results usually demands extra manual effort. This thesis investigates how human–AI collaboration can support programmers across four key stages of the workflow. To address these challenges, I begin by studying the needs and practices of programmers to understand where current tools fall short. Based on these insights, I design interactive systems that integrate with common tools such as computational notebooks and IDEs and operate on invariant components (code cells, execution outputs, text) so results remain compatible with common practices. Across the four stages, these systems provide context-aware code understanding across multiple cells, purpose-driven documentation from code and its execution results for different communicative purposes, presentation slides from code and results, and real-time, multi-dimensional code evaluation and optimization support during development, with authors remaining in control to inspect, edit, and refine outputs throughout. I conduct user studies and case studies to evaluate system usability and to assess how these approaches improve programmers’ productivity, confidence, and ability to share their work.
  • Thumbnail Image
    Item type: Item ,
    Habitat Restoration Strategies for Eastern Meadowlark (Sturnella magna, L.) in Ontario
    (University of Waterloo, 2026-05-06) Atherton, Claire
    Eastern Meadowlark (Sturnella magna, L.) is an at-risk grassland bird in Ontario. S. magna is declining in part from breeding habitat loss and quality decline. Habitat restoration has been proposed as a recovery measure. Information on S. magna’s habitat preferences and current S. magna restoration initiatives are lacking in Ontario. I studied microhabitat characteristics within restored tallgrass prairie sites in Norfolk County, Ontario. No S. magna were observed on the study sites. I compared microhabitat characteristics between potential nest attempt periods, between sites, and to time since disturbance. Visual obstruction was the only characteristic that differed between nest attempts (p < 0.05). All microhabitat characteristics differed between sites (p < 0.05). All characteristics except woody vegetation cover (dCor = -0.00038, p = 0.52) showed a correlation with time since disturbance. The levels of significance for all tests were determined to be artefacts of the small sample size and do not necessarily reflect true trends. Comparisons to the literature suggest that percent grass cover may have been too low and percent total cover too high to support S. magna, but results differed and were too sparse to make meaningful comparisons. Likely not enough time has passed since restoration for the sites to become suitable for S. magna. I distributed an online questionnaire to 334 people knowledgeable of S. magna, tallgrass prairies, and/or grassland birds about past, current, and future restoration strategies in Ontario. Thirty-five responses were received. Projects have occurred across southern Ontario with clusters near Windsor and in Northumberland County.Delayed hay harvesting was the most common management strategy in restored areas. About half of respondents indicated that post-restoration monitoring occurs at least some of thetime. Sixty-three percent of respondentsindicatedthat projects used interventions.Most respondents indicated that projects lacked in sufficiency and effectiveness. Limiting factors included finances and maintenance, and strengths included having a broad focus, planning, and monitoring. Key targets for future projects were southern and eastern Ontario. Key targets for future research were a better understanding of S. magna’s habitat needs and lifecycle, responses to restoration, and use of anthropogenic grasslands. This thesis provides an overview of the state ofS. magnarestoration in Ontario. It provides a roadmap for restoration ecologists and conservation biologists to use when managing habitat for grassland birds in Ontario.