Concussions occur at an unacceptable rate in the sport of ice hockey. Efforts are made to improve its prevention by modifying protective equipment and implementing rules of conduct; yet the effectiveness of these methods remains unknown as there is a lack of evidence pointing to a mechanical metric able to adequately predict concussion. The purpose of this thesis was to identify metrics which best characterize concussion following ice hockey collisions and provide values reflecting concussion risk. The first study reported effective mass of shoulder checks, extended elbow strikes, and tucked-in elbow strikes using 15 competitive ice hockey players as subjects. The results were used to guide the impact mass and compliance of laboratory reconstructions of real-world ice hockey collisions done in the second study. Analysis of these reconstruction showed that concussions following shoulder and elbow to head collisions in ice hockey occurred at low peak linear and angular accelerations and that impulse duration played a large role in the mechanism of injury. The results also indicated that concussion risk estimations are specific to the mechanism of loading. A 50% likelihood of concussion following a shoulder check to the head was established for peak angular accelerations of 9.2, 6.9, 4.6, and 2.2 krad/s^2 for impulse durations of 15, 20, 25, and 30 ms, respectively. A 50% likelihood of concussion following an extended elbow to the head was established for peak linear accelerations of 23, 15, and 7 g for impulse durations of 15, 20, and 25 ms, respectively. Finally, the third study reported brain tissue stress and strain comparable to the ones obtained reconstructing concussive impacts in American football, rugby, and Australian rules football despite having lower peak linear and angular acceleration values. This thesis has provided a new sport concussion data set acquired using a methodology guided by the biomechanics of ice hockey player volunteer testing, has identified metrics which can adequately predict concussion, and has established concussion risk levels. This information will be of use to helmet manufacturing companies, companies developing concussion detection sensors, and governing bodies in their efforts to eliminate concussion from the sport of ice hockey.

The ATLAS experiment at the LHC is searching for evidence of charged Higgs bosons, without yet finding it. This report investigates the possibility of reconstructing charged Higgs bosons using multi-jet final states (pp -> [b]tH+ -> [b]bbbql\nu) with focus on fully hadronic H^+ -> hW decays using a simulation of 500k events with m_{H+} = 250 GeV. After optimizing TMVA (Toolkit for Multivariate Analysis) input variables for BDT (Boosted Decision Trees), the mass of H+ was recreated with an efficiency of up to 26,7% (correct combinations).

Obtenir une vision précise des évènements survenus durant un incident est un objectif difficile à atteindre lors d'enquêtes de criminalistique informatique. Le problème de la reconstruction d'évènements, ayant pour objectif la construction et la compréhension d'une chronologie décrivant un incident, est l'une des étapes les plus importantes du processus d'investigation. La caractérisation et la compréhension complète d'un incident nécessite d'une part d'associer à chaque fragment d'information sa signification passée, puis d'établir des liens sémantiques entre ces fragments. Ces tâches nécessitent l'exploration de grands volumes de données hétérogènes trouvés dans la scène de crime. Face à ces masses d'informations, les enquêteurs rencontrent des problèmes de surcharge cognitive les amenant à commettre des erreurs ou à omettre des informations pouvant avoir une forte valeur ajoutée pour les progrès de l'enquête. De plus, tout résultat produit au terme de la reconstruction d'évènements doit respecter un certain nombre de critères afin de pouvoir être utilisé lors du procès. Les enquêteurs doivent notamment être en capacité d'expliquer les résultats produits. Afin d'aider les enquêteurs face à ces problèmes, cette thèse introduit l'approche SADFC. L'objectif principal de cette approche est de fournir aux enquêteurs des outils les aidant à restituer la sémantique des entités composant la scène de crime et à comprendre les relations liant ces entités tout en respectant les contraintes juridiques. Pour atteindre cet objectif, SADFC est composé de deux éléments. Tout d'abord, SADFC s'appuie sur des fondations théoriques garantissant la crédibilité des résultats produits par les outils via une définition formelle et rigoureuse des processus utilisés. Cette approche propose ensuite une architecture centrée sur une ontologie pour modéliser les connaissances inhérentes à la scène de crime et assister l'enquêteur dans l'analyse de ces connaissances. La pertinence et l'efficacité de ces outils sont démontrées au travers d'une étude relatant un cas d'investigation fictive. / Having a clear view of events that occurred over time is a difficult objective to achieve in digital investigations (DI). Event reconstruction, which allows investigators to build and to understand the timeline of an incident, is one of the most important steps of a DI process. The complete understanding of an incident and its circumstances requires on the one hand to associate each piece of information to its meaning, and on the other hand to identify semantic relationships between these fragments. This complex task requires the exploration of a large and heterogeneous amount of information found on the crime scene. Therefore, investigators encounter cognitive overload problems when processing this data, causing them to make mistakes or omit information that could have a high added value for the progress of the investigation. In addition, any result produced by the reconstruction process must meet several legal requirements to be admissible at trial, including the ability to explain how the results were produced. To help the investigators to deal with these problems, this thesis introduces a semantic-based approach called SADFC. The main objective of this approach is to provide investigators with tools to help them find the meaning of the entities composing the crime scene and understand the relationships linking these entities, while respecting the legal requirements. To achieve this goal, SADFC is composed of two elements. First, SADFC is based on theoretical foundations, ensuring the credibility of the results produced by the tools via a formal and rigorous definition of the processes used. This approach then proposes an architecture centered on an ontology to model and structure the knowledge inherent to an incident and to assist the investigator in the analysis of this knowledge. The relevance and the effectiveness of this architecture are demonstrated through a case study describing a fictitious investigation.

The PANDA experiment at FAIR is an antiproton experiment and it will offer unique possibilities for the study of hadron physics including hyperon physics. With the PANDA detector, it will be possible to exclusively reconstruct hyperon-antihyperon events. Hyperons are relatively long-lived particles (10−10s), which can travel for a distance of up to a few meters before decaying into their reconstructible decay products. In this project, I focus on the reaction p̅ p → Ξ̄+ Ξ− → Λ̅ π+ Λ π− → p̅ π+π+ p π− π−. The Ξ− (cascade) hyperons take a special role as they are the lightest hyperons to introduce a sequential decay topology, i.e., Ξ− → Λ π− → p π− π− (and c.c.). This constitutes a substantial challenge for the reconstruction algorithms. In this thesis, a Monte-Carlo simulation study of realistic track and event reconstruction of this Ξ̄+ Ξ− reaction in the region near the production threshold is done, and the event generation of the Ξ̄+ Ξ− reaction channel was performed by the EvtGen event generator in PandaRoot. For various realistic tracking algorithms, track and event reconstruction performance levels were investigated in PandaRoot.

L’expérience SuperNEMO cherche à observer la double désintégration beta sans émission de neutrinos, uniquement possible si le neutrino est une particule de Majorana. Le premier module, aussi appelé démonstrateur, est en cours de construction au Laboratoire Souterrain de Modane. Sa capacité à détecter les particules individuelles en plus d’en mesurer l’énergie en fait un détecteur unique. Le démonstrateur peut contenir 7 kg de ⁸²Se sous forme de fines feuilles source. Ces feuilles source sont entourées par une chambre à fils, permettant ainsi la reconstruction en 3 dimensions des traces de particules chargées. Un calorimètre segmenté, composé de scintillateurs plastiques couplés à des photomultiplicateurs, assure quant à lui la mesure de l’énergie de chaque particule. De plus, la chambre à fils peut être soumise à un champ magnétique afin d’identifier la charge des particules. SuperNEMO est donc capable d’effectuer la reconstruction complète de la cinématique d’un événement ainsi que d’identifier la nature des particules impliquées dans ce dernier : électrons, positrons, particules α ou encore particules γ. En pratique, la reconstruction des particules repose sur divers algorithmes implémentés dans un logiciel de simulation et de reconstruction développé par et pour la collaboration SuperNEMO. La reconstruction des particules γ est particulièrement délicate puisque ces particules ne laissent pas de traces dans la chambre à fils et sont seulement détectées par le calorimètre, parfois même plusieurs fois. Différentes approches ont été explorées durant cette thèse. Ce travail a abouti à la création d’un nouvel algorithme permettant à la fois d’optimiser l’efficacité de reconstruction des particules γ mais aussi d’améliorer la reconstruction de leurs énergies. D'autres programmes assurant l’identification des particules et l’opération des mesures topologiques pertinentes à chaque événement ont aussi été développés. La valeur du champ magnétique a été optimisée pour la recherche de la désintégration 0νββ à l’aide de simulations Monte-Carlo. Les performances des blindages magnétiques ainsi que leur influence sur le champ magnétique ont été évaluées via des mesures effectuées grâce à des bobines magnétiques à échelle réduite. Le démonstrateur SuperNEMO est capable de mesurer ses propres contaminations en bruits de fond grâce à des canaux d’analyse dédiés. À l’issue d’une première prise de données de 2,5 ans, les activités visées pour les principaux bruits de fond devraient être connues précisément. En outre, la demi-vie du processus 2νββ pour le ⁸²Se devrait être mesurée avec une incertitude totale de 0,3 %.À la différence d’autres expériences double beta se basant uniquement sur la somme en énergie des deux électrons, SuperNEMO a accès à la totalité de la cinématique d’un événement et donc à de plus nombreuses informations topologiques. Une analyse multivariée reposant sur des arbres de décision boostés permet ainsi une amélioration d’au moins 10 % de la sensibilité pour la recherche de la désintégration 0νββ. Après 2,5 ans, et si aucun excès d'événements 0νββ n'est observé, le démonstrateur pourra établir une limite inférieure sur la demi-vie du processus 0νββ : T > 5.85 10²⁴ ans, équivalant à une limite supérieure sur la masse effective du neutrino mββ < 0.2 − 0.55 eV. En extrapolant ce résultat à une exposition de 500 kg.an, ces mêmes limites deviendraient T > 10²⁶ ans et mββ < 40 − 110 meV. / SuperNEMO is an experiment looking for the neutrinoless double beta decay in an effort to unveil the Majorana nature of the neutrino. The first module, called the demonstrator, is under construction and commissioning in the Laboratoire Souterrain de Modane. Its unique design combines tracking and calorimetry techniques. The demonstrator can study 7 kg of ⁸²Se, shaped in thin source foils. These source foils are surrounded by a wire chamber, thus allowing a 3-dimensional reconstruction of the charged particles tracks. The individual particles energies are then measured by a segmented calorimeter, composed of plastic scintillators coupled with photomultipliers. A magnetic field can be applied to the tracking volume in order to identify the charge of the particles. SuperNEMO is thus able to perform a full reconstruction of the events kinematics and to identify the nature of the particles involved: electrons, positrons, α particles or γ particles. In practice, the particle and event reconstruction relies on a variety of algorithms, implemented in the dedicated SuperNEMO simulation and reconstruction software. The γ reconstruction is particularly challenging since γ particles do not leave tracks in the wire chamber and are only detected by the calorimeter, sometimes multiple times. Several γ reconstruction approaches were explored during this thesis. This work lead to the creation of a new algorithm optimizing the γ reconstruction efficiency and improving the γ energy reconstruction. Other programs allowing the particle identification and performing the topological measurements relevant to an event were also developed. The value of the magnetic field was optimized for the 0νββ decay search, based on Monte-Carlo simulations. The magnetic shieldings performances and their impact on the shape of the magnetic field were estimated with measurements performed on small scale magnetic coils. The SuperNEMO demonstrator is able to measure its own background contamination thanks to dedicated analysis channels. At the end of the first 2.5 years data taking phase, the main backgrounds target activities should be measured accurately. The ⁸²Se 2νββ half-life should be known with a 0.3 % total uncertainty. Unlike other double beta decay experiments relying solely on the two electrons energy sum, SuperNEMO has access to the full events kinematics and thus to more topological information. A multivariate analysis based on Boosted Decision Trees was shown to guarantee at least a 10 % increase of the sensitivity of the 0νββ decay search. After 2.5 years, and if no excess of 0νββ events is observed, the SuperNEMO demonstrator should be able to set a limit on the 0νββ half-life of T > 5.85 10²⁴ y, translating into a limit on the effective Majorana neutrino mass mββ < 0.2 − 0.55 eV. Extrapolating this result to the full-scale SuperNEMO experiment, i.e. 500 kg.y, the sensitivity would be raised to T > 10²⁶ y or mββ < 40 − 110 meV.

I detta examensarbete har tidsstämplar extraherats ur ett forensiskt perspektiv från ett hemautomationssystem med styrenheten Homey från Athom. Först konstruerades ett fiktivt händelsescenario gällande ett inbrott i en lägenhet med ett hemautomationssystem. Hemautomationssystemet bestod av flera perifera enheter som använde olika trådlösa nätverksprotokoll. Enheterna triggades under händelsescenariot. Därefter testades olika metoder för att få ut data i form av tidsstämplar. De metoder som testades var rest-API, UART och chip-off på flashminnet medan JTAG inte hanns med på grund av tidsbrist. Den metod som gav bäst resultat var rest-API:t som möjliggjorde extrahering av alla tidsstämplar samt information om alla enheter. I flashminnet hittades alla tidsstämplar, men det var inte möjligt att koppla ihop dessa tidsstämplar med en specifik enhet utan att använda information från rest-API:t. Trots att rest-API:t gav bäst resultat så var det den metod som krävde en mängd förutsättningar i form av bland annat inloggningsuppgifter eller en rootad mobil. Med hjälp av de extraherade tidsstämplarna rekonstruerades sedan händelsescenariot för inbrottet. / The purpose of this bachelor thesis was to extract timestamps from a home automation system with a control unit named Homey in a forensic perspective. The first step was to create a course of event regarding a burglar breaking into an apartment with home automation. The home automation system consisted of some peripheral units using different types of wireless network protocols. All these units were triggered during the break in. Thereafter different types of methods were tested in an attempt to extract the timestamps for each unit. These methods included rest-API, UART and chip-off on a flash memory. The method using JTAG were not tested due to lack of time. Rest-API was the method that provided most information about the units and time stamps. The flash memory also contained every timestamp, however it did not provide any information about which timestamp belonged to which unit. Even though the rest-API was the best method to extract data, it was also the method with most requirements such as credentials or a rooted smartphone. With the extracted timestamps it was possible to reconstruct the course of events of the break-in.