News

Excellent Doctoral Thesis Awards 2022 in science

  • Faculté des Sciences, des Technologies et de Médecine (FSTM)
    24 février 2023
  • Catégorie
    Recherche
  • Thème
    Ingénierie, Mathématiques, Physique & sciences des matériaux, Sciences de la vie & médecine

The Doctoral School of Science and Engineering (DSSE) at the University of Luxembourg has recently awarded 13 doctoral candidates for their outstanding doctoral thesis. Their research covers a wide variety of topics, tackling important issues and providing innovative solutions. They now continue their career in both academia and industry.

INVESTIGATING RENEWABLE ENERGIES

Christina Schubert

Thesis title: Evaluation von Synergieeffekten zentraler Speichersysteme in Niederspannungsnetzen durch integrative Modellbildung

Doctoral programme: Engineering Sciences

Current position: Product Manager Energy Storage at J.M. VOITH SE & Co. KG.

In her thesis, Christina focuses on the acceptance and integration of storage technologies by addressing the needs of and benefits for grid operators. She designed a simulation model based on standardised input data which can be accessed from smart meters by every grid operator. While renewable energies are clearly the future technology to decarbonise our energy system, their dynamics can cause discrepancies between generation and consumption, leading to more efforts in order to balance the grid. She presented an innovative concept for developing decision criteria for storage placement, dimensioning, and operation.

INVESTIGATING AI, CRYPTOGRAPHY, DRONES, MACHINE LEARNING AND SECURITY

Alessio Buscemi

Thesis title: Automation of controller area network reverse engineering: approaches, opportunities, and security threats

Doctoral programme: Engineering Sciences

Current position: Postdoctoral Researcher, SECAN-Lab, University of Luxembourg

His thesis investigates the possibility of employing artificial intelligence to automate the reverse engineering of the Controller Area Network (CAN). CAN is present in all commercial cars and trucks throughout the world and enables the communication between their electronic sensors. The CAN data provides a significant source of information that automotive aftermarket companies can use for a variety of business purposes. The dissertation proposes unique AI-based pipelines aimed at revealing these formats via a systematic analysis of the retrieved raw data from the vehicles. Furthermore, this work addresses potential cybersecurity vulnerabilities coming from the high level of automation achieved by the presented approaches.

Salah Ghamizi

Thesis title: Multi-objective robust machine learning for critical systems with scarce data

Doctoral programme: Computer Science and Computer Engineering 

Current position: Postdoctoral Researcher, Serval Group, University of Luxembourg

While research in machine learning and artificial intelligence assumes we have access to large loads of data, real-world critical systems that use machine learning face on the contrary a scarcity of useful data. My research focused on 3 topics: pandemic forecasting in Luxembourg, credit scoring and fraud detection for BGL-BNP Paribas, chest X-ray automated diagnosis. His thesis introduces a new framework (Knowledge Augmentation) with multiple algorithms and theorems that significantly improves the performance of machine learning models without the need of collecting more data to train them.

Christof Ferreira Torres

Thesis title: From smart to secure contracts: automated security assessment and improvement of Ethereum smart contracts

Doctoral programme: Computer Science and Computer Engineering 

Current position: Postdoctoral Researcher, Secure & Trustworthy Systems Group at ETH Zurich

In collaboration with Spuerkeess and the Technical University of Munich, Christof developed several techniques that would help developers write safer smart contracts. Just like traditional software, smart contracts are subject to programming mistakes. However, unlike traditional software, smart contracts are public and immutable, meaning that their code can be seen by anyone but cannot be changed once it has been deployed. Combined with the fact that smart contracts often manage assets worth millions of euros, they make an appealing target for attackers. He developed several automated tools, which developers can use to automatically detect, and fix vulnerabilities contained in their smart contracts.

Đình Hiếu Tràn

Thesis title: 5G and beyond networks with UAV: trajectory design and resource allocation

Doctoral programme: Computer Science and Computer Engineering 

Current position: Senior research specialist at Nokia Paris, France

The key goal of his PhD was to propose and develop new frameworks and efficient optimisation algorithms to solve novel challenging problems, facilitate the design and deployment of drones in wireless communications. These proposed algorithms or frameworks can become one of the foundations for deploying drones in 6G wireless systems. The main contribution of this dissertation is to propose effective methods to overcome limitations in disasters, energy saving, and limited battery lifetime.

Giuseppe Vitto

Thesis title: Security, scalability and privacy in applied cryptography

Doctoral programme: Computer Science and Computer Engineering

Current position: Researcher 

In his thesis, Giuseppe investigated the security, scalability and privacy properties of existing cryptosystems and presented novel designs and cryptanalytic results regarding blockchain primitives and public-key schemes based on elliptic curves. In his works, he studied accumulators, where he cryptanalyzed previous schemes and proposed a new scalable and privacy-friendly design for anonymous/verifiable credential systems, and he reported cryptanalytical results on primitives adopted or considered for adoption in top blockchain-based cryptocurrencies. He then focused on the study of public-key primitives based on elliptic curves, providing a method for distributed generation of semiprimes not requiring a statistical semiprimality test, and describing attack optimizations and specific implementation design choices that allowed to break a reduced-parameters instance, proposed by Microsoft, of SIKE, a post-quantum key-encapsulation mechanism candidate in the NIST Post-Quantum Standardization process.

INVESTIGATING MATERIALS

Emanuele Penocchio

Thesis title: Thermodynamics of chemical engines: a chemical reaction network approach

Doctoral programme: Physics and Materials Science

Current position: Postdoctoral researcher at Northwestern University

Chemical processes in closed systems inevitably relax to equilibrium. Energy can be employed to counteract such tendency and drive reactions against their spontaneous direction. This nonequilibrium driving is implemented in open systems, which living organisms provide the most spectacular examples of. In recent years, experiments in supramolecular chemistry, photochemistry and electrochemistry demonstrated that, by opening synthetic systems to matter and/or energy exchanges with the environment, artificial systems with life-like behaviours can be realised and used to convert energy inputs of different nature into work at both the nanoscopic and the macroscopic level. However, one tool that is still lacking is a firm grasp of the thermodynamics of these chemical engines. In this thesis, Emanuele provides it by leveraging the most recent developments of the thermodynamic description of deterministic chemical reaction networks.

Mohit Sood

Thesis title: Interface Open-Circuit Voltage Deficit in Cu(In,Ga)S2 Solar Cell: Characterization, Simulation and Mitigation

Doctoral programme: Physics and Materials Science

The origin of interface recombination in Cu(In,Ga)S2 solar cells prepared with Cu‑excess films has perplexed researchers for decades as the classical model used to explain interface recombination doesn’t agree with the experimental observations. His doctoral thesis purposes an alternate model comprising near interface defects instead of classical phenomena to explain the baffling cause of interface recombination in these devices. Using numerical simulations, he exhibited that this model can reproduce all the experimental characteristics observed for the Cu‑excess device, thus asserting the validity of the model. Further, he devised a device structure with an alternate buffer and window layer to reduce interface recombination that resulted in power conversion efficiencies on par with record Cu(In,Ga)S2 solar cell.

Alvar Torelló Massana

Thesis title: Electrocaloric coolers and pyroelectric energy harvesters based on multilayer capacitors of Pb(Sc0.5Ta0.5)O3

Doctoral programme: Physics and Materials Science

Current position: Postdoctoral researcher, Universitat Politecnica de Catalunya

He studied a novel cooling technology called electrocaloric cooling. It is based on solid-state materials (ceramics and polymers) that have the capability to change their temperature when they are connected to a battery.  He developed novel electrocaloric demonstrators and pushed their performance close to the real values that real applications demand, which had never been done. He had crossed key milestones imposed by the cooling industry that make us be very optimistic about the future of this technology. This opens a new range of applications of electrocaloric materials with a very bright future as well. 

INVESTIGATING ANTIMICROBIAL RESISTANCE, BREAST CANCER, INTESTINAL INFLAMMATION

Lynn Bonetti

Thesis title: The Th17 cell – IL22 axis depends on glutathione upon intestinal inflammation

Doctoral programme: Systems and Molecular Biomedicine

During her PhD, Lynn investigated the influence of the antioxidant glutathione in Th17 cells. She was able to decipher a signaling pathway in Th17 cells that depends on glutathione and leads to the formation of the protective cytokine IL-22. Using the murine bacterial infection model Citrobacter rodentium, she could show that glutathione in T cells is essential for the production of IL-22 and this T cell-generated IL-22 is in turn important in maintaining the integrity of intestinal tissue to prevent bacterial spread.

Arnaud Mazier

Thesis title: Data-driven patient-specific breast modeling: a simple, automatized, and robust computational pipeline

Doctoral programme: Computational Sciences 

Current position: R&D machine learning engineer at USEDGE 

Breast cancer impacts one woman over 8 in addition to a yearly increasing incidence rate. Tumors are localised and identified using magnetic resonance imaging (MRI) in the prone position, breast hanging downwards, while for practical reasons, tumor resection is performed in the supine pose, patient laid on the back. This requires an additional complex, invasive, and unsafe procedure to mark the tumor with a metallic harpoon or radioactive tags to keep track of the lesion during the surgery. Consequently, developing a numerical method to predict the tumor movement between the imaging and operative configuration is of significant interest. In this PhD project, a numerical simulation pipeline allowing the prediction of patient-specific breast tumor movement was put forward, including personalised preoperative surgical drawings.

Laura de Nies

Thesis title: Microbiome reservoirs of antimicrobial resistance

Doctoral programme: Systems and Molecular Biomedicine

Current position: Postdoctoral Research Associate, University of Oxford

Antimicrobial resistance (AMR) presents a global threat to public health due to the inability to comprehensively treat bacterial infections. Emerging resistant bacteria residing within human, animal and environmental reservoirs may spread from one to the other, at both local and global levels. Consequently, AMR has the potential to rapidly become pandemic whereby it is no longer constrained by either geographical or human-animal borders. To identify antimicrobial resistance genes and compare their identity and prevalence across different microbial reservoirs, she provided a comprehensive assessment of the prevalence of antimicrobial resistance and its dissemination mechanisms in human, animal, and environmental mechanisms.

— 

INVESTIGATING HYPERBOLIC SURFACES

Nhât Minh Doàn

Thesis title: Quantifying some properties of curves and arcs on hyperbolic surfaces

Doctoral programme: Mathematics and Applications

Current position: Researcher at the Department of Topology and Geometry, Hanoi Institute of Mathematics

His thesis focuses on the applications of the ergodicity of geodesic flows on hyperbolic surfaces. He delt with finding upper bounds for the length of a certain type of curve which is epsilon-dense on these surfaces. Then, he investigated the terms in the Luo-Tan identity, looked at how the lengths of a certain family of closed simple curves can be used to compare two different hyperbolic surfaces, and provided an upper bound on the number of specific types of subsurfaces on the surface. He examined the structure of certain types of arcs on hyperbolic surfaces and how they can provide information about the properties of the hyperbolic surface.

— 

More information about the Doctoral School in Science and Engineering (DSSE): https://dsse.uni.lu