Chalmers Open Digital Repository
Not a member yet
26247 research outputs found
Sort by
Leveraging Machine Learning in CFAR Detectors
No full textWithin the field of radar detection, a property called constant false alarm rate (CFAR) is of great importance. This property is used when, for example, creating generalized likelihood ratio test(GLRT)-based CFAR adaptive detectors. By using the statistics of these detectors, the authors of the article CFAR Feature Plane: A Novel Framework for the Analysis and Design of Radar Detectors have mapped radar data to a two dimensional feature space, called CFAR feature plane. During this work, this mapping chain was used to map data of high dimension to the plane, making it possible to compare clusters that forms in the feature plane. For larger steering vectors as input data (dimension of evaluated cell), there were less characteristics present in terms of amplitude and rotation between clusters of targets and clutter.
Furthermore, multiple estimations of the covariance matrix were used, both in terms of the amount of samples, but also together with regularization techniques, such as diagonal loading and fast maximum likelihood estimation. What could be seen was that the greater amount of data used for estimation, or by utilizing one of the regularization techniques, the more distinct clusters where formed in the CFAR feature
plane, making the final classification easier. The main goal of this project was to implement different machine learning algorithms, trained in the feature plane, to investigate if it was possible to get a more robust detector in terms of mismatched targets, than the traditional model based ones, such as Kelly’s detector.
Characteristics for the clusters, how each cluster is distributed in feature space, directly affect detector performance. The larger the steering vector, the more intersection between clusters, causing the trained machine learning detectors to adapt to the behavior of a Kelly’s detector, performing relatively well. As for the smaller sizes of input data, it is possible to create machine learning detectors that performs better than the traditional ones, both in terms of perfect matched and mismatched targets. Such algorithms are multilayered perceptrons and symbolic classifiers
AI-Powered Behavioral Analysis of Vehicle Communication to Strengthen API Security
No full textAs vehicles become increasingly connected, the volume of API communication between cars and cloud-based services grows, exposing new security risks. Traditional rule-based security systems, such as AWS Web Application Firewall, are limited to detecting known threats and patterns that can be pre-defined in the ruleset. This thesis explores the use of AI-powered anomaly detection, specifically the Isolation Forest algorithm, as a complement to existing rule-based methods to secure API traffic in connected vehicles. A series of experiments were conducted using both synthetic and real-world API request data. The results show that Isolation Forest can effectively detect anomalous requests, especially when trained on sufficiently large and representative datasets. Comparisons with a rule-based system revealed that AI-based methods might be better at identifying unknown threats, while rulebased filters remain reliable for known attack patterns. Overall, the study highlights the potential of combining machine learning with traditional approaches to create more adaptive and intelligent API security systems for connected vehicles
Konceptutveckling av visuellt underlag till högtalare
No full textProduktklimatet bland dagens portabla Bluetooth-högtalare domineras av produkter i mindre skala. Med den lilla storleken följer att ljudet från högtalarna är därefter. Det svenska konsultbolaget ESSIQ vill utforska detta produktklimat utifrån en befintlig, mycket större, produkt de tycker har ett oöverträffat ljud, vilket skall användas som projektets referens. Vad som efterfrågas är ett nytt högtalarkoncept som skall bygga på referensproduktens styrkor och adressera dess svagheter.
Denna rapport ämnar att redovisa processen kring att ta fram ett visuellt underlag, ett koncept som utgångsläge för vidare iteration och utveckling. Arbetet inleds med att undersöka vad marknaden har att erbjuda för olika sorters portabla högtalare, för att ge perspektiv till hur konceptet kan positionera sig i förhållande till konkurrenterna. Utifrån denna förstudie skall följande huvudfrågor besvaras,
Till vad, när och för vem skall konceptet utformas och hur skall detta kommuniceras i form?
Vad som resulterar är ett konceptförslag ämnad att fungera likväl inne som ute, oavsett miljöns krav och påfrestningar, av en vid målgrupp. Processen beskriver utförligt hur detta förverkligas utifrån analyser, personas, scenarion, funktionsbeskrivningar, materialval och formspråk
Evaluation of Different Additives in a Polyolefin System and Their Impact on the Polymer Properties
No full textThe effect of various liquid and solid additives on the physical properties of a
heterophasic PP copolymer were studied. Two sample preparation methods,
compression molding and tape extrusion were used to study the mechanical and
thermal properties of the polymer. To investigate the material properties, Wide
Angle X-ray Scattering (WAXS), Small Angle X-ray Scattering (SAXS), Dynamic
Mechanical Thermal Analysis (DMTA), Tensile test, Transient Plane Source
(TPS) were performed. The incorporation of additives in polymer samples did not
exhibit a considerable influence on the thermal properties of blends; however it
affected the mechanical properties. The presence of liquid additives in the
interlamellar distance may result in an increase in the interlamellar distance
without a notable impact on lamellar thickness, as laterally investigated and
confirmed by SAXS results. The liquid in amorphous phase decreases the modulus
of the sample. Increasing interlamellar distance influenced the plastic deformation
mechanism by decreasing yield stress. The additives exhibited a plasticizing effect
on the polymer, reducing modulus in the testing temperature range, leading to an
improvement in mechanical properties without compromising the thermal stability
of the polymer blend. The additives did not affect the degree of crystallinity,
despite the changes in kinetics of crystallization. The introduction of additives to
the polymer did not cause a significant change in the degree of crystallinity;
however, a higher cooling rate in the Tape extrusion sample preparation method
led to crystallographic variations and a reduction in the content of γ- form
Analytical investigation for application of Hollow CLT elements
No full textThe construction sector today are responsible for a large portion of the global carbon
emissions, which leads to a higher demand for making the industry more sustainable.
With these increasing demands put a higher pressure on reducing the raw material
used in construction. Timber is in comparison with others a more sustainable material
because it is relatively eco-friendly.
This thesis aims to investigate the application and extent of where and when it is
appropriate to implement air gaps into the cross layers of CLT elements in relation
to structural and hygrothermal performance, as well as how the evaluation should
be structured to acquire the most volume efficient layout. The optimal arrangement
for enhancing both structural and environmental performance was determined by
methodically evaluating important elements like heat transmission, moisture transport,
and material interactions.
The investigation of structural performance showed that air gaps could be applied to
a meaningful extent in most situations and still satisfy structural requirements saving
approximately 20% of the material usage, where even with constrained widths of the
air gaps, a significant reduction of the volume in relation to its solid counterpart.
For the Hygrothermal performance no major impact could be observed from the air
gaps other than the thermal and moisture storage capacity decreased
Development of a Parametric Life Cycle Assessment (LCA) method for Construction Logistics in a Circular Economy
No full textThis thesis presents a parametric Life Cycle Assessment (LCA) model for evaluating
the environmental impacts of construction logistics in circular economy contexts. The
focus lies on quantifying the greenhouse gas emissions associated with transporting five
categories of recovered construction material such as steel reinforcement bars, glass
partitions, wooden doors, wooden flooring, and mineral wool in Sweden.Using
openLCA 2.1 and the ecoinvent 3.8 database, logistics scenarios are modeled for diesel,
biodiesel, and electric heavy goods vehicles. The analysis includes both single-load and
co-loaded transport configurations. Results show that electric trucks, despite lower
payload capacities, yield the lowest emissions due to the low-carbon Swedish electricity
mix. Co-loading strategies also prove effective in reducing the total number of trips and
overall emissions.Break-even distances are calculated to determine the threshold at
which transporting reused materials becomes environmentally beneficial, with
distances varying significantly by material and transport mode. An uncertainty analysis
was performed to evaluate the sensitivity of break-even distances under different
transport and loading conditions. The proposed model demonstrates the significance of
logistics choices in circular construction planning. It offers practical insights for
stakeholders aiming to reduce transport emissions and supports improved integration
of logistics parameters into LCA frameworks
Assessing Privacy vs. Efficiency Tradeoffs in Open-Source Large Language Models
No full textLLMs are being actively implemented across various industries in applications from customer service to code generation. With this recent development, concerns surrounding data privacy have become increasingly urgent. While open-source LLMs
are often seen as a more transparent and flexible alternative to proprietary models, the extent of their openness and privacy guarantees vary significantly, as well as the research done in this area being quite small. With regulatory pressure from the EU AI Act, many organizations must now navigate the trade-offs between transparency, privacy, and efficiency. This thesis investigates two key questions, “What are the actual privacy guarantees provided by open-source LLMs?” and “Does ensuring robust privacy safeguards in open-source LLMs necessarily compromise efficiency?”. Through our evaluation process, we find no consistent link between a model’s openness and its resistance to privacy attacks, and neither do privacy safeguards necessarily reduce efficiency. These findings suggest that it is possible to develop or select open-source models that are both privacy-conscious and efficient
Modelling and Comparative Analysis of Switchable Battery Configurations for Dual Voltage Charging in Electric Vehicles
No full textAbstract
This thesis explores and evaluates two electric vehicle (EV) battery architectures: an existing 800V single battery system using inverter-based 400V charging referred to as the DC boost system, and a proposed dual 400V battery system that allows seriesparallel switching for direct 400V charging, referred to as the Greenfield system. The main aim is to identify energy losses in these systems and investigate architectural and material solutions to enhance efficiency, thermal performance, reliability, and cost-effectiveness. A MATLAB Simulink model was created to simulate energy flows, battery behaviour, and thermal dynamics under a variety of charging conditions, spanning from realistic to sensitivity-driven scenarios. Battery balancing mechanisms were incorporated to evaluate system performance when SOC mismatches occur. Although both systems achieve comparable efficiencies during 400V charging, the proposed architecture shows lower energy losses, improved thermal performance and longer component lifespan due to decreased current stress and the removal of voltage boosting with the inverter. Furthermore, the possibility of replacing aluminium with copper busbars was studied, revealing lower thermal peaks while retaining electrical performance.
Overall, the findings indicate that even though the energy efficiency differences between the two systems are slight, the Greenfield system provides significant benefits in thermal stability and component durability, particularly during high-current operation. These results highlight the trade-offs involved in system complexity, material choice, and long-term performance in the design of EV battery systems