Innovative Projects Realized

Hierarchical Knowledge Distillation and Pruning (HKDP) Framework for Secure and Sustainable Federated Learning

The Hierarchical Knowledge Distillation and Pruning (HKDP) framework project we propose addresses the growing demand for real-time data processing in cybersecurity applications. Traditional centralized machine learning approaches are hindered by high computational requirements and privacy concerns, but federated learning (FL) offers a solution by enabling decentralized model training, maintaining data privacy, reducing energy consumption, privacy, and security assurance.
This research project aims to design, implement, and evaluate the HKDP framework for optimizing energy efficiency in FL for cybersecurity. The objective is to enhance real-time data processing while minimizing computational burdens on devices. To achieve this, the project will pursue four key sub-objectives: conducting a thorough literature review, designing the framework, implementing the HKDP framework, and experimentally evaluating its performance.
The project will investigate the impact of hierarchical knowledge distillation on energy efficiency, optimal pruning strategies, and the adaptability of energy-aware FL frameworks. Utilizing established cybersecurity datasets and PyTorch/TensorFlow implementation, evaluation metrics will include energy consumption, model accuracy, and computational requirements.
The anticipated outcomes include improved energy efficiency (up to 30% reduction), enhanced model accuracy (up to 5% improvement), and reduced computational requirements (up to 20% decrease). The HKDP framework will provide an innovative solution for energy-efficient federated learning in cybersecurity.

Designing an AI solution for data analysis to improve the process of measuring thermal properties of materials: liquids, solids, powders and pastes

One of the busines processes of C-Therm Technologies is related to devices measuring thermal properties of materials: liquids, solids, powders and pastes. One part of the process is analyzing the data collected by the sensors. The project in this proposal will involve designing an AI solution for data analysis to improve the process of collected data analysis. The analysis which is currently performed by experts, will be performed by employing statistic analysis or machine and deep learning methods. The project will address the problems of implementing prototype solutions, testing their performance and potentially integration into the company’s software.
The benefits for the partner company, involve improved productivity through automation: the experts who are currently performing a repetetive data analysis procedure, will only need to verify the suggestion system analysis.

Venture & Investment Strategist Co-op

Edmonton Community Foundation (ECF), part of a comprehensive network of Canadian community foundations, acts as a bridge between donors and greater Edmonton charities to help create and maintain a strong, vibrant community for generations to come. Through permanent endowment funds, the Foundation disburses 3.5 per cent of its assets annually to greater Edmonton charities. As well, the Foundation takes a direct role in selected community initiatives that help further the vision of a strong community.?Since inception, the Foundation has allocated more than $400 million to enhance and strengthen our community. ?

Within this commitment to community development lies the Social Enterprise Fund (SEF), a key initiative that plays a crucial role through impact investing. SEF has a mandate to achieve both social and financial goals through its investments. Finding and screening organizations that can achieve both requires complex due diligence processes, something which SEF still struggles to do successfully.

SEF would like to bring onboard social-minded and investment-focused?interns to help develop, implement and test key areas of the due diligence and tracking processes. This includes researching best practices in due diligence optimization (possibly incorporating AI), social impact analysis and impact reporting. Developing a framework to assess applicants as well as report on social impact will be critical to the SEF operations moving forward. We envision this working happening over two different periods. The first term will have the student research and conduct a pilot. In a subsequent term, the next intern will take learnings from the pilot to further refine processes in the three areas.

Tread with Caution: tracking real-world locomotor biomechanics to understand trips, slips, and falls

Millions of falls occur every year placing substantial burden on all sectors of our society. What causes these falls is difficult to identify and is made more challenging as most of the research did not occur in real-world environments. Our project will develop wearable technology and new methods to study the factors that lead to falls, be it our own movement patterns or the environment. The results of this research will not only support improvements to building code and walkway safety but directly aid MEA Forensic Engineers and Scientists in their ability to provide forensic analysis of injurious falls.

Exploring Uncertainties and Enhancing Reliability in Flight Booking Systems with NDC and Real-Time Data Integration

This project represents an initial exploration of the New Distribution Capability (NDC) standard and its potential for transforming flight booking systems. By integrating real-time flight data, the project aims to investigate the challenges, uncertainties, and opportunities associated with adopting NDC in travel agency workflows. Through a methodical approach, the initiative will identify critical barriers, evaluate initial solutions, and establish a foundation for future development. The findings from this phase will provide valuable insights into optimising operational efficiency and enhancing user satisfaction, laying the groundwork for more advanced implementations and wider adoption of NDC in subsequent stages.

Developing a spaceborne-assisted Digital Twin for sustainable mining operations

Mining sustainability as one of Canada’s strategic industries can be effectively enhanced by using digital twins meaning to consider socioeconomic impacts, conserve the natural resources for the current and future generations, and be resilient against climate change over the mine life cycle. A digital twin is a virtual replica of a physical object, system, or process that uses real-time data to simulate how it behaves in the real world. Digital twins are one of the most trending technologies that can be used for various purposes, including aerospace, architecture, construction, energy, infrastructures, manufacturing, equipment, smart cities, urban planning, etc. The current project involves three interns with backgrounds in mining and geology, AI in geotechnics, and marketing and business development in developing a platform for spaceborne-assisted digital twin for mining systems. The project will involve creating a 3D geological BIM model (building information modelling) of the mine; collection, classification and implementation of semantic data of the material, and operation; collection, processing, and integration of satellite and drone imageries; implementation of real-time data via IoT (internet of things) sensors; performing AI-assisted and InSAR analysis on spaceborne imageries; integration of all into a powerful Digital Twin of the mining system; and performing a range of business development and marketing activities related to the developed product.

Developing long-COVID biomarkers in a Pan-Canadian cohort

Long COVID (LC) is a long-lasting condition that affects people after a COVID-19 infection. It includes symptoms that can last for months and affect different parts of the body. In Canada, about 19% of adults who had COVID-19 develop LC, with many experiencing symptoms for over six months or even longer. The exact cause of LC is still unclear, leading to challenges in diagnosing and treating it.
Research suggests that LC might be caused by factors like lingering virus, reactivation of other viruses, immune system problems, and changes in the body’s blood vessels. Our early study in young paramedics found signs that the immune system may play a big role in LC. We aim to develop tests to better diagnose LC by studying samples from a large group of Canadian COVID-19 patients.

Monitoring the mental health of Canadians

The COVID-19 pandemic had a great impact on the mental health of Canadians. Surprisingly, as of April 2022, self-rated levels of anxiety and depression were comparable to the ones observed during the worst periods of the pandemic, according to data collected by MHRC. Therefore, it is important that we continue to monitor the mental health needs and strengths of the general population in Canada. The goal of this project is to establish population-based mental health metrics within a long-term mental health surveillance system to monitor the mental health needs of the population and to identify indicators critical for expanding service delivery in Canada. In addition to risk factors of symptoms of anxiety, depression and psychological distress, this project will explore substance use, post-traumatic stress disorders, social determinants of mental health, and house and food insecurity, which have not been a focus of prior MHRC analytics and reporting. Moreover, we will conduct a detailed analysis of the use of mental health services and initiate qualitative surveys to better understand the impact of the pandemic among equity-deserving populations. Data collection will be conducted on an ongoing basis by MHRC through online surveys every three months for the next three years.

Development and Testing of a Thermoelectric-Integrated Furnace for Residential Power Generation

The project aims to develop and test thermoelectric technology with residential furnaces to generate power. The solution will generate power from waste heat from furnaces with improved energy efficiency. The testing includes thermal measurements, power measurements, and energy efficiency evaluation and optimization.

Modélisation et gestion des flux dans les garages de camions lourds électriques

La transition des opérateurs de flottes de véhicules lourds vers des Véhicules à Zéro Émission (VZE) est un processus ambitieux, mais nécessaire pour atteindre les objectifs climatiques ciblés. Elle repose sur un ensemble de facteurs interdépendants : technologies adaptées, infrastructures de recharge, soutien gouvernemental, et engagement des opérateurs privés. Dans un secteur compétitif aux marges serrées et traditionnellement averse au risque, les gestionnaires des flottes privées doivent s’assurer que ce changement majeur est économiquement viable, prendre des décisions technologiques éclairées, renforcer la résilience opérationnelle et planifier financièrement la transition.
CIMA+ soutient les opérateurs privés dans leur transition vers des VZE en offrant des services de conception et d’ingénierie. Son défi principal est le manque d’outils permettant d’anticiper les impacts opérationnels et financiers des choix technologiques d’infrastructures et de modèles d’exploitation. Pour réduire les risques et augmenter les chances de succès de cette transition, CIMA+ souhaite dans le cadre de ce projet développer un outil flexible et évolutif capable de modéliser les infrastructures de recharge et les flux opérationnels dans les garages et d’expérimenter différentes scénarios. Cet outil permettra aux opérateurs d’optimiser le design d’infrastructures et leurs opérations en tenant compte des effets à long terme.

Development of High-Performance Carbon Fibers from Asphaltenes

The goal of this project is to develop effective strategies to improve the quality of CarboMat’s asphaltene-derived carbon fibers for various multifunctional composite applications. Through this project, the partner organization, CarboMat Inc., would be able to leverage their funding because of the grant from Mitacs Elevate to work towards further development of their existing asphaltene-derived carbon fiber production technology while utilizing a wide variety of expertise and resources from the University of Calgary.

Migration strategy variation among and within Strait of Georgia Chinook salmon populations

Chinook salmon exhibit diverse migration strategies, both among and with populations. In the Strait of Georgia between mainland British Columbia and Vancouver Island, Chinook salmon exhibit two distinct migration strategies. Some individuals (called migrants) migrate away to the outer coast of British Columbia, Washington and Alaska, while others (called residents) remain near their home rivers in the Strait of Georgia. This project’s goal is to track Chinook salmon to understand which individuals and populations are more likely to stay resident in the Strait of Georgia, using two methods. First, we are tracking the migrations of 329 juvenile Chinook salmon, using small electronic tags, to understand if traits like body size or growth rates influence where salmon migrate. Second, we are using extensive datasets of when and where each population is caught by fisheries and scientific surveys in the Strait of Georgia, using genetic baselines to identify fish to their population. These two studies will further our understanding of the drivers of migratory behaviour in salmon. Broadly, they will provide key context to Pacific Salmon Foundation’s ongoing work to quantify survival of Chinook salmon in the Strait of Georgia, as migrations determine the opportunities and risks that salmon experience at sea.