The demand for electricity is rising rapidly, with more power generated through re-newable energy sources. Renewable energy sources can fluctuate in their power output atshort notice, making it more difficult to maintain the balance between electricity consump-tion and production in the short term. A solution that has gained increased interest recentlyis to connect battery energy storage systems to the grid as a means of maintaining balance.However, such systems are often controlled remotely by a cloud control system, creatingtime-critical control loops over the internet that are partly responsible for the stability andcontinued function of the electrical grid. Cyberattacks against these closed-loop systemscould devastate the electrical grid and the apparatus connected to it.In this thesis, a reference model is designed for an electrical grid load-balancing cloudcontrol system connected to remote battery energy storage systems and remote grid fre-quency sensors (measuring the balance between production and consumption). The modelis evaluated from a cybersecurity perspective by implementing a simulator and applyingdifferent cyberattacks on the simulated system.The results show that some of the most critical attack methods that a threat actor couldutilize are: disrupting the connections over the internet that are part of the closed-loopsystem, abusing remote access links from the outside to gain access to subsystems (suchas seizing control of batteries), or disturbing external dependencies to the cloud such asdomain name system (DNS) and network time protocol (NTP) servers or the contractsrelated to electricity trading. The most important cyberattacks identified in the thesis are:time delay switch (delays of messages), denial of service (disturbing message availability),false data injection (modifying message contents), replay (replaying old messages), andload altering (affecting the grid balance through direct altering of electricity consumptionand production).The simulated cyberattacks differ in how they affect the grid frequency, i.e. the gridproduction-consumption balance. Large enough network packet delays caused oscilla-tions in the simulated frequency. Denial of service attacks caused unpredictable behavior,and a high enough packet drop rate caused oscillations. For false data injection, the re-sults depend on which internet link was attacked and what injection strategy was used;some attacks caused oscillations, while others caused a steady state error or even an in-creasingly deviating frequency. Replay attacks were able to cause a deviation during thereplay window when used effectively. Finally, large enough load altering caused oscilla-tions, especially when an attacker had control over at least 15% of the system’s balancingpower.Overall, attacks on the simulated system are serious and precautions must be carefullyconsidered before such a system is implemented in the real world.

Among the various natural calamities, flood is considered one of the most catastrophic natural hazards, which has a significant impact on the socio-economic lifeline of a country. The Assessment of flood risks facilitates taking appropriate measures to reduce the consequences of flooding. The flood risk assessment requires Big data which are coming from different sources, such as sensors, social media, and organizations. However, these data sources contain various types of uncertainties because of the presence of incomplete and inaccurate information. This paper presents a Belief rule-based expert system (BRBES) which is developed in Big data platform to assess flood risk in real time. The system processes extremely large dataset by integrating BRBES with Apache Spark while a web-based interface has developed allowing the visualization of flood risk in real time. Since the integrated BRBES employs knowledge driven learning mechanism, it has been compared with other data-driven learning mechanisms to determine the reliability in assessing flood risk. Integrated BRBES produces reliable results comparing from the other data-driven approaches. Data for the expert system has been collected targeting different case study areas from Bangladesh to validate the integrated system.

Bindings are used to make a software library accessible in languages other than those that the library was originally written for. However, creating and maintaining large amounts of bindings for every library is time-consuming and costly. An alternative approach to bringing functionality to more languages is to use a service-oriented architecture, where functionality is provided as services accessible from another process through message passing. Various middlewares exist to enable message passing between processes. In this thesis, some of the state of the art messaging middlewares are explored and evaluated them in terms of various criteria. Emphasis is given to their suitability for programmable built for industrial vehicles. Three of the most suitable middlewares are used to implement small systems based on a service-oriented architecture for further evaluation. The results indicate that the Data Distribution Service is the most promising candidate, owing to its interface description language, language support, and relatively low RAM and disk space usage. / Bindings används för att göra ett mjukvarubibliotek tillgängliga i andra språk än de som biblioteket till en början var gjort för. Att skapa och underhålla bindings för varje bibliotek är dock tidskrävande och kostsamt. Ett alternativt sätt att tillhandahålla funktionalitet till fler språk är att använda en tjänsteorienterad arkitektur där funktionalitet finns tillgänglig i tjänster som andra processer använder via meddelandeöverföring. Det finns flera mellanprogramvaror för meddelandeöverföring mellan processer. I denna uppsats undersöks några av de främsta mellanprogramvarorna i förhållande till en mängd kriterier, med fokus på hur lämpliga de är för programmerbara displays gjorda för industriella fordon. För ytterligare utvärdering används de tre mest lämpliga mellanprogramvarorna för att implementera små system baserade på en tjänstorienterad arkitektur. Resultaten tyder på att Data Distribution Service är den mest lovande kandidaten tack vare dess Interface Description Language, språkstöd och relativt låga användning av RAM och diskutrymme.

This thesis explores the integration of generative AI and large language models (LLMs) into software documentation processes, assessing their impact on developer productivity. The research focuses on the development of a documentation system powered by an LLM, which automates the creation and retrieval of software documentation. The study employs a controlled experiment followed by a survey involving software developers to quantify changes in productivity through various metrics such as effectiveness, velocity, and quality of documentation generated by the system. Background: The increasing complexity of software development necessitates efficient documentation systems. Traditional methods, often manual and time-consuming, struggle to keep pace with the dynamics of software development, potentially leading to outdated and inadequate documentation. Objectives: To investigate whether a documentation system powered by an LLM can enhance developers’ productivity in software documentation tasks by assisting developers with the creation of development documentation and facilitating the retrieval of relevant information. Method: A controlled experiment followed by a survey were conducted, wherein participants were tasked with generating and using documentation through both manual and LLM-assisted methods. The effectiveness, velocity, and quality of the documentation were measured and compared. Results: The findings indicate that the LLM-powered documentation system significantly enhances developer productivity. Developers using the system were able to produce and comprehend documentation more quickly and accurately than those using the manual method. Furthermore, the quality of the documentation, assessed in terms of comprehensibility, completeness, and readability, was consistently higher when generated by the LLM system. Conclusions: The integration of LLMs into software documentation processes can significantly enhance developer productivity by automating routine tasks and improving the quality of documentation. This supports software developers in maintaining current projects and also assists in the onboarding process of new team members by providing easier access to necessary documentation. / Denna avhandling utforskar integrationen av generativ AI och stora språkmodeller (LLM) i processer för mjukvarudokumentation, och bedömer deras inverkan på utvecklares produktivitet. Forskningen fokuserar på utvecklingen av ett dokumentationssystem drivet av en LLM, som automatiserar skapandet och hämtningen av mjukvarudokumentation. Studien använder ett kontrollerat experiment följt av en enkät som involverar professionella mjukvaruutvecklare för att kvantifiera förändringar i produktivitet genom olika mått som effektivitet, hastighet och kvalitet på dokumentation genererad av systemet. Bakgrund: Den ökande komplexiteten i mjukvaruutveckling kräver effektiva dokumentationssystem. Traditionella metoder, ofta manuella och tidskrävande, har svårt att hålla jämna steg med dynamiken i mjukvaruutveckling, vilket potentiellt kan leda till föråldrad och otillräcklig dokumentation. Syfte: Att undersöka om ett dokumentationssystem drivet av en LLM kan förbättra utvecklares produktivitet i uppgifter relaterade till mjukvarudokumentation genom att assistera utvecklare med att skapa utvecklingsdokumentation och underlätta hämtningen av relevant information. Metod: Ett kontrollerat experiment följt av en enkät genomfördes, där deltagarna hade i uppgift att generera och använda dokumentation genom både manuella och LLM-assisterade metoder. Effektiviteten, hastigheten och kvaliteten på dokumentationen mättes och jämfördes. Resultat: Resultaten visar att dokumentationssystemet drivet av LLM väsentligen förbättrar utvecklarnas produktivitet. Utvecklare som använde systemet kunde producera och förstå dokumentation snabbare och mer exakt än de som använde den manuella metoden. Vidare var kvaliteten på dokumentationen, bedömd i termer av begriplighet, fullständighet och läsbarhet, konsekvent högre när den genererades av LLM-systemet. Slutsatser: Integrationen av LLM i mjukvarudokumentationsprocesser kan väsentligen förbättra utvecklarnas produktivitet genom att automatisera rutinuppgifter och förbättra kvaliteten på dokumentation. Detta stöder inte bara mjukvaruutvecklare i att underhålla pågående projekt utan hjälper också till med introduktionen av nya teammedlemmar genom att ge enklare tillgång till nödvändig dokumentation.

This thesis investigates the computing capabilities of three distinct platforms for radio direction finding (RDF) applications in electronic warfare (EW) systems: the Raspberry Pi 4 Model B, Intel NUC NUC7i5BNH, and NVIDIA Jetson AGX Orin 64GB. RDF plays a critical role in locating radio emitters, demanding real-time processing for precise signal data analysis. The study aims to determine the maximum sampling frequency that each platform can maintain while meeting real-time requirements and identifies the most suitable RDF algorithm for platform assessment. The best-suited algorithm was found to be Phase Interferometry. Results indicate that the Raspberry Pi 4 Model B achieves a sampling frequency of 13.08 MHz, the Intel NUC NUC7i5BNH maintains 12.68 MHz, and the NVIDIA Jetson AGX Orin 64GB performs at 399.45 MHz (60W), 229.82 MHz (50W), 83.88 MHz (30W), and 54.12 MHz (15W).

No description available.

This master’s thesis investigates the feasibility of producing synthetic Standard Definitionmap data using Generative AI techniques in collaboration with Repli5 AB. The study commences with a comprehensive literature review to establish context and insights into prior work in the field. Subsequently, it engages with the company’s provided training software, focusing on creating Standard Definition Maps through Generative AI. The methodology encompasses a series of procedures utilizing the company’s already existing training infrastructure to fine-tune the generative models in order to generate standard definition maps. The iterative nature of this process is geared towards enhancing existing road masks with urban features. Following familiarization with training procedures and initial experimentation, the focus shifts to devising evaluation methods for generated images. These techniques encompass training a model capable of generating realistic maps, manual annotation of images in several criteria, and training multiple evaluation architectures to identify the most effective one. By exploring the intersection of Generative AI and map data generation, this research aims to advance both theoretical understanding and practical applications in various industries, such as the autonomous vehicle sector. The ability to generate realistic urban structures has the potential to facilitate vehicle simulation and enhance safety testing protocols, with a significant aspect being the evaluation of these generated maps.

In the continuously evolving field of malware investigation, a Trojan horse, which appears as innocent software from the user's perspective, represents a significant threat and challenge for antivirus solutions because of their deceptive nature and the various malicious functionalities they provide. This study will compare the effectiveness of three free antiviruses for Linux systems (DrWeb, ClamAV, ESET NOD32) against a dataset of 1919 Trojan malware samples. The evaluation will assess their detection capabilities, resource usage, and the core functionalities they offer. The results revealed a trade-off between these three aspects: DrWeb achieved the highest detection rate (93.43%) but consumed the most resources and provided the most comprehensive functionalities. While ClamAV balanced detection and resource usage with less functionality, ESET NOD32 prioritised low resource usage but showcased a lower detection rate than the other engines (80.93%). Interestingly, the results showed that the category of Trojan horse malware and the file format analysed can affect the detection capabilities of the evaluated antiviruses. This suggests that there is no “silver bullet” for Linux systems against Trojans, and further research in this area is needed to assess the detection capabilities of antivirus engines thoroughly and propose advanced detection methods for robust protection against Trojans on Linux systems.

Modern indoor positioning systems encounter challenges inherent to indoor environments. Signal changes can stem from various factors like object movement, signal propagation, or obstructed line of sight. This thesis explores a supervised machine learning approach that integrates Bluetooth Low Energy (BLE) and inertial sensor data to achieve consistent angle and distance estimations. The method relies on BLE angle estimations and signal strength alongside additional sensor data from an Inertial Measurement Unit (IMU). Relevant features are extracted and a supervised learning model is trained and then validated on familiar environment tests. The model is then gradually introduced to more unfamiliar test environments, and its performance is evaluated and compared accordingly. This thesis project was conducted at the u-blox office and presents a comprehensive methodology utilizing their existing hardware. Several extensive experiments were conducted, refining both data collection procedures and experimental setups. This iterative approach facilitated the improvement of the supervised learning model, resulting in a proposed model architecture based on transformers and convolutional layers. The provided methodology encompasses the entire process, from data collection to the evaluation of the proposed supervised learning model, enabling direct comparisons with existing angle estimation solutions employed at u-blox. The results of these comparisons demonstrate more accurate outcomes compared to existing solutions when validated in familiar environments. However, performance gradually declines when introduced to a new environment, encountering a wider range of signal conditions than the supervised model had trained on. Distance estimations are then compared with the path loss propagation equation, showing an overall improvement. / Moderna inomhuspositioneringssystem möter utmaningar som förekommer i inomhusmiljöer. Signalförändringar kan bero på olika faktorer som objektets rörelse, signalutbredning eller blockerad siktlinje. Denna kandidat avhandling undersöker ett övervakat maskininlärningssätt som integrerar Bluetooth Low Energy (BLE) och tröghetssensorer för att uppnå konsekventa vinkel- och avståndsberäkningar. Metoden bygger på BLE-vinkelberäkningar och signalstyrka tillsammans med ytterligare sensordata från en Inertial Measurment Unit (IMU). Relevanta funktioner extraheras och en övervakad inlärningsmodell tränas och valideras sedan på tester i bekanta miljöer. Modellen introduceras sedan gradvis till mer obekanta testmiljöer, och dess prestanda utvärderas och jämförs därefter. Detta examensarbete genomfördes på u-blox kontor och presenterar en omfattande metodik som utnyttjar deras befintliga hårdvara. Flera omfattande experiment genomfördes, vilket förfinade både datainsamlingsprocedurer och experimentuppsättningar. Detta iterativa tillvägagångssätt underlättade förbättringen av den övervakade inlärningsmodellen, vilket resulterade i en föreslagen modellarkitektur baserad på transformatorer och konvolutionella lager. Den tillhandahållna metodiken omfattar hela processen, från datainsamling till utvärdering av den föreslagna övervakade inlärningsmodellen, vilket möjliggör direkta jämförelser med befintliga vinkelberäkningslösningar som används på u-blox. Resultaten av dessa jämförelser visar mer exakta resultat jämfört med befintliga lösningar när de valideras i bekanta miljöer. Dock minskar prestandan gradvis när den introduceras till en ny miljö, där den möter ett bredare spektrum av signalförhållanden än vad inlärningsmodellen har tränats på. Avståndsberäkningar jämförs sedan med en matematisk formel, kallat path loss propagation ekvationen, som ger distans som en funktion av uppmätt signalstyrka.

This thesis explores the use of machine learning (ML) and smartphone sensors to improve indoordistance estimation, a critical aspect of healthcare tests like the 6-minute walk test (6MWT). In order to make tests like the 6MWT more available, and lower the barrier for a patient toget tested, there are multiple problems which need to be solved: How can the distance data needed for these tests be collected reliably and remotely, and without having to rely on the patient reporting correct data; How can these tests be performed indoors, without relying on GPS or other GNSS, which are unreliable indoors. To tackle these challenges a convolutional neural network (CNN) trained on a dataset containing continuous ground truth was employed. An enhancement of an existing CNN model was done by collecting more training data, tuning hyper parameters, and testing it on a diverse dataset. The results of this thesis shows that when predicting distance walked on data from participants the CNN model has seen before, the precision meets clinical minimum for being able to show a change in the health condition. On real world data the performance suffers. Despite limitations due to the scope of data collection, the results still underscores the potential of ML for accurate and efficient indoor distance estimation and points to future research directions. / <p></p><p></p><p></p>