Le problème de gestion de projet à contraintes de ressources est un des problèmesles plus étudiés dans la littérature. Il consiste à planifier des activités soumises à desrelations de précédence, et nécessitant des ressources renouvelables. L’objectif est deminimiser la durée du projet, soit le makespan. Nous étudions le problème de gestion deprojet à contraintes de ressources. Nous nous sommes intéressées à la résolution exactedu problème. Dans la première partie de la thèse, nous élaborons une série de bornesinférieures basées sur le raisonnement énergétique et des formulations mathématiques.Les résultats montrent que les bornes proposées surpassent ceux de la littérature. Dansla deuxième partie, nous proposons des procédures par séparation et évaluation utilisantles bornes inférieures dévelopées dans la première partie. / Resource Constrained Project Scheduling Problem is one of the most studied schedulingproblems in the literature. It consists in scheduling activities, submitted to precedencerelationship, and requiring renewable resources to be processed. The objective isto minimize the project duration, i.e., the makespan. We study the Resource ConstrainedProject Scheduling Problem. We are interested on the exact resolution of the problem.In the first part of the thesis, we develop a series of lower bounds based on energeticreasoning and mathematical formulations. The computational results show that theproposed lower bounds outperform the ones of the literature. In the second part, wepropose Branch-and-Bound procedures using the lower bounds developed on the firstpart.

Abstract The concept of the Internet of Things (IoT) is driven by advancements of the Internet with the interconnection of heterogeneous smart objects using different networking and communication technologies. Among many underlying networking technologies for the IoT, Wireless Sensor Network (WSN) technology has become an integral building block. IoT enabled sensor networks provide a wide range of application areas such as smart homes, connected healthcare, smart cities and various solutions for the manufacturing industry. The integration of WSNs in IoT will also create new security challenges for establishing secure channels between low power sensor nodes and Internet hosts. This will lead to many challenges in designing new key establishment and authentication protocols and redefining the existing ones. This dissertation addresses how to integrate lightweight key management and authentication solutions in the resource constrained sensor networks deployed in IoT domains. Firstly, this thesis elaborates how to exploit the implicit certificates to initiate secure End-to-End (E2E) communication channels between the resource constrained sensor nodes in IoT networks. Implicit certificates are used for authentication and key establishment purposes. The compliance of the security schemes is proven through performance evaluations and by discussing the security properties. Secondly, this dissertation presents the design of two lightweight group key establishment protocols for securing group communications between resource-constrained IoT devices. Finally, the thesis explores promising approaches on how to tailor the existing security protocols in accordance with IoT device and network characteristics. In particular, variants of Host Identity Protocol (HIP) are adopted for constructing dynamic and secure E2E connections between the heterogeneous network devices with imbalanced resource profiles and less or no previous knowledge about each other. A solutions called Collaborative HIP (CHIP) is proposed with an efficient key establishment component for the high resource-constrained devices on the IoT. The applicability of the keying mechanism is demonstrated with the implementation and the performance measurements results. / Tiivistelmä Esineiden internet (IoT) on viime aikoina yleistynyt konsepti älykkäiden objektien (smart objects) liittämiseksi internetiin käyttämällä erilaisia verkko- ja kommunikaatioteknologioita. Olennaisimpia esineiden internetin pohjalla toimivia teknologioita ovat langattomat sensoriverkot (WSN), jotka ovat esineiden internetin perusrakennuspalikoita. Esineiden internetiin kytketyt langattomat sensoriverkot mahdollistavat laajan joukon erilaisia sovelluksia, kuten älykodit, etäterveydenhuollon, älykkäät kaupungit sekä älykkäät teollisuuden sovellukset. Langattomien sensoriverkkojen ja esineiden internetin yhdistäminen tuo mukanaan myös tietoturvaan liittyviä haasteita, sillä laskentateholtaan yleensä heikot anturit ja toimilaitteet eivät kykene kovin vaativiin tietoturvaoperaatioihin, joihin lukeutuvat mm. tietoturva-avaimen muodostus ja käyttäjäntunnistus. Tässä väitöskirjassa pyritään vastaamaan haasteeseen käyttämällä kevyitä avaimenmuodostus- ja käyttäjäntunnistusratkaisuja esineiden internetiin kytketyissä resurssirajoitetuissa sensoriverkoissa. Väitöstutkimuksessa keskitytään aluksi implisiittisten sertifikaattien käyttöön tietoturvallisten end-to-end-kommunikaatiokanavien alustamisessa resurssirajoitettujen sensori- ja muiden IoT-laitteiden välillä. Implisiittisiä sertifikaatteja käytetään käyttäjäntunnistuksessa sekä avaimenmuodostuksessa. Kehitettyjen ratkaisujen soveltuvuus tarkoitukseen osoitetaan suorituskykymittauksilla sekä vertaamalla niiden tietoturvaomi- naisuuksia. Seuraavaksi väitöskirjassa esitellään kaksi kevyttä ryhmäavaimenmuodostus- protokollaa tietoturvalliseen ryhmäkommunikaatioon resurssirajoitettujen IoT-laitteiden välillä. Lopuksi väitöskirjassa tarkastellaan lupaavia lähestymistapoja olemassa olevien tietoturvaprotokollien räätäläintiin IoT-laitteiden ja -verkkojen ominaisuuksille sopiviksi. Erityistä huomiota kiinnitetään Host Identity -protokollan (HIP) eri versioiden käyttöön dynaamisten ja tietoturvallisten end-to-end-yhteyksien luomiseen toisilleen ennestään tuntemattomien erityyppisten IoT-laitteiden välillä, joiden laitteistoresurssiprofiilit voivat olla hyvin erilaiset. Väitöskirjan keskeinen tulos on väitöskirjatyössä kehitetty Colla- borative HIP (CHIP) -protokolla, joka on resurssitehokas avaimenmuodostusteknologia resurssirajoitetuille IoT-laitteille. Kehitetyn teknologian soveltuvuutta tarkoitukseensa demonstroidaan prototyyppitoteutuksella tehtyjen suorituskykymittausten avulla.

La complexité croissante du matériel et du logiciel dans le développement des applications pour les systèmes embarqués induit de nouveaux besoins et de nouvelles contraintes en termes de fonctionnalités, de capacité de stockage, de calcul et de consommation d'énergie. Cela entraîne des difficultés accrues pour les concepteurs alors que les contraintes commerciales liées aux développement et à la production de ces systèmes demeurent impondérables. Un autre défi qui s'ajoute à cette complexité est le développement des applications avec de fortes exigences de sécurité et de fiabilité (S&D) pour des systèmes embarqués contraints en ressources (RCES). De ce fait, nous recommandons d'aborder cette complexité via la réutilisation d'un ensemble d'artefacts de modélisation dédiés. Le "patron" constitue l'artefact de base pour représenter des solutions S&D, à spécifier par les experts de ces aspects et à réutiliser par les ingénieurs pour résoudre les problèmes de l'ingénierie système/logicielle du domaine confrontée à ces aspects. Dans le cadre de cette thèse, nous proposons une approche d'ingénierie à base de modèles pour la spécification, le packaging et la réutilisation de ces artefacts pour modéliser et analyser ces systèmes. Le fondement de notre approche est un ensemble de langages de modélisation et des règles de transformation couplés à un référentiel orienté modèles et de moteurs de recherche et d'instantiation. Ces langages de modélisation permettent de spécifier les patrons, les systèmes de patrons et un ensemble de modèles de propriétés et de ressources. Ces modèles permettent de gouverner l'utilisation des patrons, leur organisation et leur analyse pour d'éventuel réutilisation. Pour les transformations, nous avons conçu un ensemble de règles pour l'analyse de la satisfiabilité des architectures logicielles à base de patrons S&D pour des plateformes matérielles à base de modèles de ressources. Pour le développement du référentiel, nous proposons un processus de spécification et de génération basé sur les langages de modélisation décrits au préalable. Les moteurs permettent de retrouver pour ensuite instantier ces artefacts vers des environnements de développement spécifiques.Dans le cadre de l'assistance pour le développement des applications S&D pour les RCES, nous avons implémenté une suite d'outils, basés sur Eclipse EMF/QVTO, pour supporter la spécification de ces artefacts et l'analyse des applications S&D autour d'un référentiel. Afin de faciliter l'utilisation de l'approche proposée, nous préconisons un ensemble de méthodes pour favoriser l'utilisation de cette suite d'outils tout au long du processus de développement. En outre, nous avons abordé la construction par génération automatique de référentiels orientés modèles, basée sur Eclipse EMF/QVTO et la plateforme Eclispe CDO, accompagné d'un ensemble d'outils pour le peuplement, l'accès et la gestion de ce référentiel. Les problèmes étudiés dans ce travail ont été identifiés dans le contexte du projet européen FP7 TERESA. Les solutions proposées ont ainsi été démontrées et évaluées dans le cadre de ce projet à travers un cas d'étude d'une application de contrôle, exhibant des exigences de fiabilité et de sécurité, pour les systèmes ferroviaires. / Non-functional requirements such as Security and Dependability (S&D) become more and more important as well as more and more difficult to achieve, particularly in embedded systems development. Such systems come with a large number of common characteristics, including real-time and temperature constraints, security and dependability as well as efficiency requirements. In particular, the development of Resource Constrained Embedded Systems (RCES) has to address constraints regarding memory, computational processing power and/or energy consumption. In this work, we propose a modeling environment which associates model-driven paradigm and a model-based repository, to support the design and the packaging of S&D patterns, resource models and their property models. The approach is based on a set of modeling languages coupled with a model-repository, search and instantiation engines towards specific development environments. These modeling languages allow to specify patterns, resources and a set of property models. These property models will allow to govern the use of patterns and their analysis for reuse. In addition, we propose a specification and generation process of repositories. As part of the assistance for the development of S&D applications, we have implemented a suite of tool-chain based on the Eclipse platform to support the different activities around the repository, including the analysis activities. The proposed solutions were evaluated in the TERESA project through a case study from the railway domain.

La plupart des travaux de recherches sur les problèmes d'ordonnancement traitent le cas des ressources renouvelables, c'est-à-dire des ressources qui sont exigées en début d'exécution de chaque tâche et sont restituées en fin d'exécution. Peu d'entre eux abordent les problèmes à ressources consommables, c'est-à-dire des ressources non restituées en fin d'exécution. Le problème de gestion de projet à contraintes de ressources (RCPSP) est le problème à ressources renouvelables le plus traité dans la littérature. Dans le cadre de cette thèse, nous nous sommes intéressés à une généralisation du problème RCPSP qui correspond au cas où les tâches sont remplacées par des événements liés par des relations de précédence étendues. Chaque événement peut produire ou consommer une quantité de ressources à sa date d'occurrence et la fonction économique reste la durée totale à minimiser. Nous avons nommé cette généralisation ERCPSP (Extended RCPSP). Nous avons élaboré des modèles de programmation linéaire pour résoudre ce problème. Nous avons proposé plusieurs bornes inférieures algorithmiques exploitant les travaux de la littérature sur les problèmes cumulatifs. Ensuite, nous avons élargi la portée des méthodes utilisées pour la mise en place de méthodes de séparation et évaluation. Nous avons traité aussi des cas particuliers par des méthodes basées sur la programmation dynamique. / This thesis investigates the Extended Resource Constrained Project Scheduling Problem (ERCPSP). ERCPSP is a general scheduling problem where the availability of a resource is depleted and replenished at the occurrence times of a set of events. It is an extension of the Resource Constrained Project Scheduling Problem (RCPSP) where activities are replaced by events, which have to be scheduled subject to generalized precedence relations. We are interested in this thesis in proposing new methodologies and approaches to solve ERCPSP. First, we study some polynomial cases of this problem and we propose a dynamic programming algorithm to solve the parallel chain case. Then, we propose lower bounds, mixed integer programming models, and a branch-and-bound method to solve ERCPSP. Finally, we develop an instance generator dedicated to this problem.

Magister Artium (Social Work) - MA(SW) / The burden of HIV disease is concentrated in sub-Saharan Africa and South Africa (SA) is particularly affected. Whilst there have been many studies conducted on the biomedical and socio-psychological aspects of HIV and AIDS, insufficient attention has been paid to the quality of life of those infected with the virus. The primary purpose of this study was to determine the predictors of quality of life enjoyment and satisfaction (Q-LES) of individuals living with HIV and or AIDS and those on anti-retroviral treatment or being prepared for it. Further, the study determined the relationship between psychological distress and Q-LES of HIV positive individuals because psychological distress is reported to contribute substantially to the burden of the disease in sub-Saharan Africa, including SA. This quantitative study used a battery of questionnaires administered to 121 participants in an out-patient clinic setting. The main hypothesis tested in this study is: psychological distress is a strong predictor of Q-LES. The Hospital Anxiety and Depression Scale (HADS) was used to screen for psychological distress and the Quality of Life Enjoyment and Satisfaction Questionnaire (Q-LES-Q) to assess the various components of QoL. A purposive sampling strategy was used to recruit participants into the study. Data analysis included descriptive and inferential statistics using SPSS to test the hypothesis. Of the total sample (N=121), 74% were females. The study found that a large proportion (49.5%) of the sample within the age group 25-49 years old had significant presence of psychological distress. Those not on ART yet were significantly affected (66%). The relationship between Q-LES subscales and psychological distress was significant (p< 0.01). The results show that psychological distress was significantly prevalent among HIV positive individuals and it was the strongest predictor of Q-LES among the study participants. Modifying the current psychological intervention programmes, in the public health clinics, for individuals vi infected with the HI virus will assist in improving the current health outcomes and also help to achieve better Quality of Life outcomes.

Vision systems have become ubiquitous. They are used for traffic monitoring, elderly care, video conferencing, virtual reality, surveillance, smart rooms, home automation, sport games analysis, industrial safety, medical care etc. In most vision systems, the data coming from the visual sensor(s) is processed before transmission in order to save communication bandwidth or achieve higher frame rates. The type of data processing needs to be chosen carefully depending on the targeted application, and taking into account the available memory, computational power, energy resources and bandwidth constraints. In this dissertation, we investigate how a vision system should be built under practical constraints. First, this system should be intelligent, such that the right data is extracted from the video source. Second, when processing video data this intelligent vision system should know its own practical limitations, and should try to achieve the best possible output result that lies within its capabilities. We study and improve a wide range of vision systems for a variety of applications, which go together with different types of constraints. First, we present a modulo-PCM-based coding algorithm for applications that demand very low complexity coding and need to preserve some of the advantageous properties of PCM coding (direct processing, random access, rate scalability). Our modulo-PCM coding scheme combines three well-known, simple, source coding strategies: PCM, binning, and interpolative coding. The encoder first analyzes the signal statistics in a very simple way. Then, based on these signal statistics, the encoder simply discards a number of bits of each image sample. The modulo-PCM decoder recovers the removed bits of each sample by using its received bits and side information which is generated by interpolating previous decoded signals. Our algorithm is especially appropriate for image coding. / Morbee, M. (2011). Optimized information processing in resource-constrained vision systems. From low-complexity coding to smart sensor networks [Tesis doctoral no publicada]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/12126 / Palancia

The adoption of small-size Unmanned Aerial Vehicles (UAVs) in the commercial and professional sectors is rapidly growing. The miniaturisation of sensors and processors, the advancements in connected edge intelligence and the exponential interest in Artificial Intelligence (AI) are boosting the affirmation of autonomous nano-size drones in the Internet of Things (IoT) ecosystem. However, achieving safe autonomous navigation and high-level tasks like exploration and surveillance with these tiny platforms is extremely challenging due to their limited resources. Lightweight and reliable solutions to this challenge are subject to ongoing research. This work focuses on enabling the autonomous flight of a pocket-size, 30-gram platform called Crazyflie in a partially known environment. We implement a modular pipeline for the safe navigation of the nano drone between waypoints. In particular, we propose an AI-aided, vision-based reactive planning method for obstacle avoidance. We deal with the constraints of the nano drone by splitting the navigation task into two parts: a deep learning-based object detector runs on external hardware while the planning algorithm is executed onboard. For designing the reactive approach, we take inspiration from existing sensorbased navigation solutions and obtain a novel method for obstacle avoidance that does not rely on distance information. In the study, we also analyse the communication aspect and the latencies involved in edge offloading. Moreover, we share insights into the finetuning of an SSD MobileNet V2 object detector on a custom dataset of low-resolution, grayscale images acquired with the drone. The results show the ability to command the drone at ∼ 8 FPS and a model performance reaching a COCO mAP of 60.8. Field experiments demonstrate the feasibility of the solution with the drone flying at a top speed of 1 m/s while steering away from an obstacle placed in an unknown position and reaching the target destination. Additionally, we study the impact of a parameter determining the strength of the avoidance action and its influence on total path length, traversal time and task completion. The outcome demonstrates the compatibility of the communication delay and the model performance with the requirements of the real-time navigation task and a successful obstacle avoidance rate reaching 100% in the best-case scenario. By exploiting the modularity of the proposed working pipeline, future work could target the improvement of the single parts and aim at a fully onboard implementation of the navigation task, pushing the boundaries of autonomous exploration with nano drones. / Användningen av små obemannade flygfarkoster (UAV) inom den kommersiella och professionella sektorn ökar snabbt. Miniatyriseringen av sensorer och processorer, framstegen inom connected edge intelligence och det exponentiella intresset för artificiell intelligens (AI) ökar användningen av autonoma drönare i nanostorlek i ekosystemet för sakernas internet (IoT). Att uppnå säker autonom navigering och uppgifter på hög nivå, som utforskning och övervakning, med dessa små plattformar är dock extremt utmanande på grund av deras begränsade resurser. Lättviktiga och tillförlitliga lösningar på denna utmaning är föremål för pågående forskning. Detta arbete fokuserar på att möjliggöra autonom flygning av en 30-grams plattform i fickformat som kallas Crazyflie i en delvis känd miljö. Vi implementerar en modulär pipeline för säker navigering av nanodrönaren mellan riktpunkter. I synnerhet föreslår vi en AI-assisterad, visionsbaserad reaktiv planeringsmetod för att undvika hinder. Vi hanterar nanodrönarens begränsningar genom att dela upp navigeringsuppgiften i två delar: en djupinlärningsbaserad objektdetektor körs på extern hårdvara medan planeringsalgoritmen exekveras ombord. För att utforma den reaktiva metoden hämtar vi inspiration från befintliga sensorbaserade navigeringslösningar och tar fram en ny metod för hinderundvikande som inte är beroende av avståndsinformation. I studien analyserar vi även kommunikationsaspekten och de svarstider som är involverade i edge offloading. Dessutom delar vi med oss av insikter om finjusteringen av en SSD MobileNet V2-objektdetektor på en skräddarsydd dataset av lågupplösta gråskalebilder som tagits med drönaren. Resultaten visar förmågan att styra drönaren med ∼ 8 FPS och en modellprestanda som når en COCO mAP på 60.8. Fältexperiment visar att lösningen är genomförbar med drönaren som flyger med en topphastighet på 1 m/s samtidigt som den styr bort från ett hinder som placerats i en okänd position och når måldestinationen. Vi studerar även effekten av en parameter som bestämmer styrkan i undvikandeåtgärden och dess påverkan på den totala väglängden, tidsåtgången och slutförandet av uppgiften. Resultatet visar att kommunikationsfördröjningen och modellens prestanda är kompatibla med kraven för realtidsnavigering och ett lyckat undvikande av hinder som i bästa fall uppgår till 100%. Genom att utnyttja modulariteten i den föreslagna arbetspipelinen kan framtida arbete inriktas på förbättring av de enskilda delarna och syfta till en helt inbyggd implementering av navigeringsuppgiften, vilket flyttar gränserna för autonom utforskning med nano-drönare.

As resource-constrained autonomous vehicles are used for more and more applications, their ability to achieve the lowest possible localization error without expending more power than needed is crucial. Despite this, the parameter settings of the localization systems, both for the platform and the application, are often set arbitrarily. In this thesis, we propose a model-based controller that adapts the parameters of the localization system during run-time by observing conditions in the environment. The test-bed used for experiments consists of maplab, a visual-inertial localization framework, that we execute on the Nvdia Jetson AGX platform. The results show that the linear velocity is the single most important environmental attribute to base the decision of when to update the parameters upon. We also found that while it was not possible to find a direct connection between certain parameters and environmental conditions, a connection could be found between sets of configuration parameters and conditions. Based on these conclusions, we compare model-based controller setups based on three different models: Finite Impulse Response (FIR), AutoRegressive eXogenous input (ARX) and Multi-Layer Perceptron (MLP). The FIR-based controller performed the best. This FIR-based controller is able to select configurations at the appropriate times to keep the error lower than it would be to randomly guess which set of configuration parameters is best. The proposed solution requires offline profiling before it can be implemented on new localization systems, but it can help to reduce the error and power consumption and thus enable more uses of resource-constrained devices. / Användningen av autonoma fordon med begränsade resurser ökar allt mer, vilket i sin tur ökar vikten av att dessa kan lokalisera med lägsta möjliga fel utan att förbruka mer effekt. Trots detta bestäms parametrarna för både hårdvara och i algoritmerna ofta godtyckligt för dessa lokaliseringssystem. I detta examensarbete presenterar vi en lösning till detta, i form av en modellbaserad regulator som anpassar parametrarna baserat på vad den detekterar i omgivningen. Vår testuppställning består av maplab, ett lokaliseringsramverk, som vi exekverar på Nvida Jetson AGX plattformen. Resultaten visar att den linjära hastigheten är den viktigaste miljövariabeln att detektera och använda för att anpassa parametrarna i lokaliseringssystemet. Resultaten visar även att det går att hitta kopplingar mellan konfigurationer och miljövariabler, även om det inte går att hitta mellan specifika konfigurationsparameterar och miljövariabler. Den regulator som presterar bäst visar sig vara en som är baserad på en Finite Impulse Response modell, med en optimeringshorisont på 5 sekunder. Denna presterar bättre än både AutoRegressive eXogenous input baserad regulator och en Multi-Layer Perceptron baserad regulator. Finite Impulse Response regulatorn åstadkommer ett fel som är lägre än slumpmässig gissning, på data den inte sett förut. Lösningen som uppvisas i detta projekt kräver optimering offline för att fungera, men om det utförs kan den reducera både lokaliseringsfelet och effektförbrukningen och genom det skapa nya användningsområden för resursbegränsade enheter.

Time series anomaly detection has been a longstanding area of research with applications across various domains. In recent years, there has been a surge of interest in applying deep learning models to this problem domain. This thesis presents a critical examination of the efficacy of deep learning models in comparison to classical approaches for time series anomaly detection. Contrary to the widespread belief in the superiority of deep learning models, our research findings suggest that their performance may be misleading and the progress illusory. Through rigorous experimentation and evaluation, we reveal that classical models outperform deep learning counterparts in various scenarios, challenging the prevailing assumptions. In addition to model performance, our study delves into the intricacies of evaluation metrics commonly employed in time series anomaly detection. We uncover how it inadvertently inflates the performance scores of models, potentially leading to misleading conclusions. By identifying and addressing these issues, our research contributes to providing valuable insights for researchers, practitioners, and decision-makers in the field of time series anomaly detection, encouraging a critical reevaluation of the role of deep learning models and the metrics used to assess their performance. / Tidsperiods avvikelsedetektering har varit ett långvarigt forskningsområde med tillämpningar inom olika områden. Under de senaste åren har det uppstått ett ökat intresse för att tillämpa djupinlärningsmodeller på detta problemområde. Denna avhandling presenterar en kritisk granskning av djupinlärningsmodellers effektivitet jämfört med klassiska metoder för tidsperiods avvikelsedetektering. I motsats till den allmänna övertygelsen om överlägsenheten hos djupinlärningsmodeller tyder våra forskningsresultat på att deras prestanda kan vara vilseledande och framsteg illusoriskt. Genom rigorös experimentell utvärdering avslöjar vi att klassiska modeller överträffar djupinlärningsalternativ i olika scenarier och därmed utmanar de rådande antagandena. Utöver modellprestanda går vår studie in på detaljerna kring utvärderings-metoder som oftast används inom tidsperiods avvikelsedetektering. Vi avslöjar hur dessa oavsiktligt överdriver modellernas prestandapoäng och kan därmed leda till vilseledande slutsatser. Genom att identifiera och åtgärda dessa problem bidrar vår forskning till att erbjuda värdefulla insikter för forskare, praktiker och beslutsfattare inom området tidsperiods avvikelsedetektering, och uppmanar till en kritisk omvärdering av djupinlärningsmodellers roll och de metoder som används för att bedöma deras prestanda.