ISEL - Eng. Elect. Tel. Comp. - Dissertações de Mestrado
URI permanente para esta coleção:
Navegar
Entradas recentes
- Standard-based smart card access control architecture for critical infrastructuresPublication . Oliveira, António Filipe Ramic Santos; Dias, Tiago Miguel Braga da Silva; Chaves, Ricardo Jorge FernandesAbstract This thesis proposes an enterprise-centered access control architecture that is technically compatible with widely adopted standards, enabling the convergence of physical and logical access on a single smart card credential while preserving interoperability and enterprise control over identity proofing, registration, and credential issuance. This need is particularly acute for operators of critical infrastructure, who must replace legacy badge-based systems, ensure cross-site and outage resilience, unify access control with support for rapid revocation, generate audit trails, and avoid vendor lock-in through the use of open standards. The proposed architecture was built to meet these needs. The main outcomes of this work are a standards-aligned enterprise architecture for a multisite access control system and a smart card credential with a well-defined, standardized data model and lifecycle. Additionally, this work showcases a proof-of-concept prototype consisting of a Card Management System that personalizes cards and issues Public Key Infrastructure (PKI) credentials. The prototype also features a Java Card applet that implements a standardized data model and command set, supporting asymmetric cryptography-based multi-factor authentication mechanisms. Furthermore, it includes an access control system emulator for testing purposes. Using this stack, it is possible to issue cards and test them by performing strong authentication mechanisms, demonstrating the end-to-end feasibility of the devised solution for both physical and logical access scenarios.
- Aplicação para caracterização da vozPublication . Santos, Pedro Branco; Cordeiro, Hugo TitoEste projeto propõe uma aplicação simples e intuitiva com a capacidade de se ligar a um servidor, com o foco a melhorar a experiência do utilizador através de dinamismo e modelação. A funcionalidade principal da aplicação é apresentar resultados, numa interface, que caracterizam a fala através da comunicação com o servidor, mantendo sempre a troca de informação segura e que não haja perdas desta. Um dos objetivos principais do projeto foi garantir que a experiência do utilizador seja agradável, removendo qualquer complexidade na interface da aplicação. A arquitetura do servidor está desenhada de modo a garantir que a informação é sempre processada e que o utilizador obtenha sempre os resultados desejados. Este projeto desenvolveu uma aplicação protótipo que permite a extração de parâmetros dos sinais de fala para rastreio de patologias da voz. A aplicação pretende ser dinâmica, nomeadamente através da inclusão de novos algoritmos. A interface de utilizador foi desenvolvida em Unity3D de modo a permitir o uso em múltiplas plataformas. Os algoritmos foram desenvolvidos em Python de modo a tirar partido das bibliotecas de processamento de fala e sinal existentes.
- Application of stereoscopy in speleological surveyingPublication . Gonçalves, Martim Augusto Teixeira; Fazenda, Pedro Viçoso; Jorge, Pedro Miguel Torres MendesAbstract Cave topography plays a fundamental role in supporting various fields that require the exploration of underground environments by specialists. However, speleologists rely on traditional techniques which remain labor-intensive and prone to errors. While LiDAR and photogrammetry boast advanced mapping accuracy, high costs, bulk, laborous preparation and operational complexity limit widespread adoption. This study aims to investigate the use of a portable, cost-effective alternative surveying method by leveraging stereoscopy and established tools from the robotics ecosystem consisting on the ZED 2 3D camera and OctoMap framework. Field experiments compared stereoscopic mapping against a traditional compass-and-laser workflow, evaluating accuracy, efficiency and usability. Results demonstrate that stereo-SLAM pipelines produce metrically accurate 3D models in real time, offering an interesting development path to bridge the gap between manual surveys and high-end LiDAR scans. Limitations in portability, environmental conditions and robustness were identified along with future directions to address them. Despite these limitations in the method, it shows promising results by reducing cost and effort in obtaining structured and machine-readable cave representations with applications beyond speleology. These findings may support the development of tools to assist adjacent fields such as archaeology, geology, biology and environmental monitoring. The work establishes a foundation fieldready stereoscopic systems supporting semantic mapping, advanced spatial analysis, and integration with robotic exploration.
- Driver profile and drowsiness classificationPublication . Valente, Duarte Faria da Mota Gonçalves; Lourenço, André Ribeiro; Ferreira, Artur JorgeAbstract Every day, approximately 3,700 people die in road accidents, totaling 1.35 million fatalities globally each year. The primary causes of these accidents include speeding, distracted driving, drunk driving, nighttime driving, and drowsy driving. Changing human driving habits is extremely challenging, as road safety campaigns alone have shown limited impact on reducing fatalities. However, while influencing behavior is challenging, technological advancements offer promising solutions to mitigate risks and enhance road safety. This thesis explores unsafe driving behaviors, with a particular focus on both dangerous and drowsy driving. Using data from in-car sensors and physiological signals, we investigate ethods to assess driver states and identify patterns associated with an increased risk of accidents. Specifically, we analyze driving behavior metrics, such as speeding, harsh braking, and sudden acceleration, to classify risky driving tendencies. Additionally, we leverage heart rate variability features extracted from electrocardiograms recorded via a sensor-equipped steering wheel to detect signs of driver drowsiness. The results indicate that both behavioral and physiological markers can serve as effective indicators of unsafe driving conditions. In particular, aggressive driving behaviors are strongly linked to accident risk, while prolonged driving and fatigue significantly impair driver performance. Moreover, individual differences in responses to sleep deprivation highlight the need for personalized assessment methods. These insights contribute to the development of intelligent monitoring systems capable of identifying and mitigating unsafe driving conditions in real time, ultimately enhancing road safety.
- Electric vehicle X driving range prediction 2 EV X DRP2Publication . Valido, João Francisco Fidalgo; Ferreira, Artur Jorge; Coutinho, David PereiraAbstract The use of Electric Vehicles (EV) has increased in recent years. The autonomy of the EV, expressed as its Driving Range (DR) is a key factor. This autonomy depends on several variables related to the vehicle itself as well as with external conditions. An accurate estimation of the DR value at each moment is a challenging task. In this thesis, we address the DR estimation problem using machine learning techniques. We build a dataset with 11 features, for DR estimation, using publicly available EV data. Then, we discuss the use of Machine Learning (ML) Regression techniques to estimate DR, with Linear Regression (LR), Multilayer Perceptron (MLP), and Radial Basis Function (RBF) neural networks. Moreover, we assess the effect of unsupervised dimensionality reduction techniques using feature selection and feature reduction approaches. The experimental results show that the use of both feature selection and feature reduction are useful at reducing the dimensionality of the data, keeping or improving the performance for DR estimation. This study also identifies the top features for DR estimation. The best feature selection method was the Mean-Median approach, while Principal Component Analysis yielded the best results in terms of feature reduction. Among the regression techniques evaluated, linear regression achieved the best overall performance. However, in real-world scenarios, where a larger number of variables may be present, methods such as MLP or RBF might offer better adaptability and robustness.
- RFID solutions evaluation in industrial container access controlPublication . Silva, Tiago Duque Leite Vieira da; Serrador, António João NunesAbstract Identifying and recording the flow of industrial waste containers through access points can pose significant challenges in industrial environments. Manual registration methods can lead to registration failures that pose risks for a company, particularly for the security of assets and operational efficiency. Thus, the identification of container crossings through the facility’s gateways is essential for reinforcing security measures and optimising operational management in industrial environments. This work investigates the application of passive RFID (Radio Frequency IDentification) technology to address these challenges. A pilot architecture was developed, the most suitable hardware was selected and tested, and the system was installed in a waste management company, supported by a background service for processing and recording tag event data. The system was tested in real conditions with various crossing scenarios. RFID tags were installed on the company’s containers, and their passages through the gateway were monitored. Detection quality and middleware performance were evaluated through transaction statistics. The results show that the passive rigid tags used in the project significantly improve their performance when they have a metal background, increasing their detection capacity from 64% to 100%. It was also concluded that a company can benefit significantly in financial terms as a result of improved operational management and the deterrent effect on theft.
- Monitoring quality of air in urban environments using LoRa technologyPublication . Balona, Bernardo da Silva; Pires, Luís Miguel RegoAbstract This dissertation main objective to monitor quality of air in urban environments which are densely populated using quality of air sensors, to ensure real-time monitoring, LoRa/LoRaWAN communication will be used and therefore studied in order to understand the communication in metropolitan areas. This dissertation aims to fully experiment with the communications of values read by sensors and a LoRa cloud, all while using LoRa and a gateway. This type of project can extrapolate and evaluate the use of LoRaWAN in smart cities. By using LoRa/LoRaWAN, the objective consists of finding a low-cost option that provides real-time monitoring to measure and acquire sensor data. Moreover, this integration will allow to evaluate the viability of LoRa/LoRaWAN networks for environmental monitoring in smart cities and as a viable option for institutions to use this solution to easily monitor air quality in dense urban cities and vast countryside. This report presents a state of art of the components the system will use such as sensors, a MCU that will employ LoRa communication and a Gateway to pass the acquired sensor data to a server cloud which will then show the values through an API, which showed expected values of concentration, percentage and other units, such as micrometers, in the atmosphere. With the dioxide carbon sensor, we could go even further in the experimentation and understood the value expected of concentration would be around 350 ppm, with the practical values being 351 ppm. Results about values read will then be presented to understand the quality of air in the studied area and to compare these results with official values and make comparisons, measuring, the sensors’ efficiency. Radio test results will also be discussed to better understand the communication length that LoRa can achieve in cities with the higher values around 1,7 km with a Spreading Factor of 7 and with a Transmitting Power of 10, 12 and 14 dBm. In addition to this entire system, it was also created a predictive system to conclude about the general quality of air with the values obtained.
- Design of a programmable photonic integrated circuit based on carrier depletion phase shiftersPublication . Velázquez, Ernesto Lazaro Chavez; Lourenço, Paulo Jorge Passos Sério; Fantoni, AlessandroAbstract Photonic integrated circuits (PICs) have appeared in response to current technological challenges such as increased speed, demand for greater bandwidth and processing capacity, especially for applications that require extremely fast, high-performance data processing that Integrated Circuits (ICs) cannot satisfy. The range of applications for this technology includes ultra-fast data centers, optical communications, biosensors, sensors for vehicles, quantum processors, and Artificial Intelligence (AI). As this technology continues to advance, there is a growing academic and scientific need to contribute to technology, which in turn opens new market opportunities. In this work, we will analyze the main optoelectronic properties of the materials used for PIC design and the latest advances in waveguide and coupler topologies, focusing on hydrogenated amorphous silicon (a-Si:H) as the main material and directional couplers as the main structure. An architecture will be proposed that looks to respond to problems in conventional directional couplers. It is also intended to use this new coupler design to create a programmable circuit that can be controlled by the Thermo-Optic Effect (TOE) and the Carrier Depletion Phenomenon (CDP). The methodology used integrates the most recent and classic studies of the mathematical and physical principles that govern these structures and materials, using analysis tools such as MATLAB and Python, with the aim of laying the theoretical foundations for the use of professional design tools. in our case, we will use Synopsys' Rsoft-CAD Layout, which implements advanced algorithms such as: Finite Difference Time Domain (FDTD), Beam Propagation Method (BPM), and Finite Element Method (FEM). In conclusion, a document will be presented that has the relevant theory of PICs and their control methods, as well as two functional solutions and their applications.
- Digital assistant with artificial intelligence techniquesPublication . Dias, Dinis Rodrigues; Ferreira, Artur Jorge; Leite, Nuno Miguel da Costa de SousaAbstract The design, implementation, and assessment of a modular Digital Assistant (DA), developed in Python, that can process natural language in speech and text, being optimized for the Windows desktop environment are presented in this dissertation. The DA performs tasks like retrieving weather data and launching applications, where the system combines Large Language Models (LLM) to interpret user requests and dynamically choose between conversational responses and function execution. To ensure modularity, extensibility, and maintainability, a layered architecture was used to organize the functionality, reasoning engine, conversation handling, and graphical user interface modules. To maintain responsiveness and user control even during lengthy operations, the assistant uses asynchronous execution, supports both text and voice input, and can output speech synthesis. The implementation places a strong emphasis on sound software engineering techniques, such as modular contracts, interface-first design, and reliable error handling. The secure handling of Application Programming Interface (API) keys and the lack of persistent memory protect privacy are also addressed. Experimental evaluation shows near real-time responses from contemporary LLM backends, sub-second latency for functionality modules, and high accuracy in differentiating between function calls and conversations. Additionally, qualitative validation verifies that the system satisfies its non-functional requirements for modularity, robustness, and user experience, and that the Graphical User Interface (GUI) is responsive and the speech features are usable. In conclusion, the project produces a useful, expandable, and intuitive digital assistant that connects conversational Artificial Intelligence (AI) and desktop task automation, providing a solid basis for upcoming improvements like cross-platform deployment, sophisticated speech recognition, and runtime model selection.
- Advanced function composition in serverless platformsPublication . Silva, Tiago Luís Lima da; Freitas, Filipe Bastos de; Simão, José Manuel de Campos Lages GarciaAbstract Serverless computing, particularly Function-as-a-Service (FaaS) platforms, allows developers to focus on the software engineering aspects of their services without managing the underlying infrastructure. These platforms rely on stateless functions that are triggered by events, making them a common choice for workflows and function composition. However, despite their advantages, serverless workflows often require developers to meet provider-specific requirements, leading to portability challenges and vendor lock-in. Previous work has attempted to address these limitations. The QuickFaaS project demonstrated the importance of standardizing function definitions across platforms to create a uniform programming model. Building on this, the OmniFlow project introduced a Domain-Specific Language (DSL) that enables developers to define serverless workflows in a provider-agnostic manner, allowing them to be reused across different cloud environments without modification. This work extends OmniFlow by introducing additional capabilities that enhance serverless workflow execution and function composition. The proposed enhancements include control flow-based workflow execution for repetitive tasks, enabling the definition of iterations within their workflows without relying on provider-specific construct. In addition, it also introduces support for parallel execution, allowing workflows to scale efficiently. By leveraging parallel processing, serverless applications can execute independent tasks concurrently, improving performance and reducing execution time. Additionally, this research explores cross-cloud function composition, ensuring that workflows can seamlessly integrate functions across multiple cloud providers, to mitigate vendor lock-in and allow developers to optimize performance by leveraging the strengths of different platforms while maintaining a unified workflow definition. The proposed enhancements provide a more flexible, scalable, and portable approach to serverless workflow orchestration, enabling developers to build complex workflows that are not constrained by the limitations of individual cloud providers.