ISEL - Eng. Elect. Tel. Comp. - Dissertações de Mestrado
Permanent URI for this collection
Browse
Recent Submissions
- Engineering the SHIDA super-app: research, design and development of a literature-centered social network with e-commerce and e-learningPublication . Ribeiro, Diogo Filipe Ricardo; Fazenda, Pedro Viçoso; Mehari, TheodrosIn the realm of software engineering and web development, the SHIDA superapp serves as a case study navigating complex challenges. This research explores software architecture, agile methodologies, and the evolving landscape of full-stack development. It scrutinizes the journey from requirement analysis to testing while addressing issues like technological debt and scope creep. By examining the SHIDA project, this study uncovers obstacles and opportunities, offering insights into code maintain ability, scalability, and performance optimization in modern software development.
- Future railway mobile communication system automated planningPublication . Queirós, Artur Daniel Rocha; Cota, Nuno; Pato, Matilde Pós-de-MinaWith the end of support for the Global System for Mobile Communications Railway (GSM-R) in sight, there needs to be a change in the technology used in railway communications. To achieve this, operators began the transition to Future Railway Mobile Communication System (FRMCS). The work developed in this thesis demonstrates the application of the concept of ge netic algorithms in a telecommunications network planning. Where the objective is to understand whether the results obtained are viable comparing to the currently practiced values. To achieve this, it is necessary to develop an algorithm that allows to obtaining the best possible solution for the placement of antennas along the line in the most efficient way, taking into account its coverage. The work developed includes the construction of this same algorithm and all its phases.Using the Cascais line as a test subject and with the help of data made available by the company Solvit, it is possible to obtain different scenarios varying four parameters, the population size, the number of generations, the crossover probability and the mutation probability. The final results prove that the use of genetic algorithms to optimize a railway telecommunications network can be a useful and powerful tool, as the results obtained presents an optimized value compared to the current solution used by public operators.
- Análise de medidas de poupança energética em redes móveis multi-tecnologiaPublication . Rodrigues, Viviana Alexandra Carrola; Vieira, Pedro Manuel de Almeida CarvalhoA crescente preocupação com a eficiência energética, sustentabilidade e aumento dos custos energéticos tem levado empresas a focarem-se na redução dos seus consumos. Por esse motivo, os operadores de redes móveis têm trabalhado para otimizar energeticamente a sua rede, tornando-se assim mais sustentáveis. Com o objetivo de reduzir os custos operacionais através da poupança energética, nesta dissertação foram propostas um conjunto de soluções, nomeadamente a modernização de equipamentos, a desativação da tecnologia de Terceira Geração (3G), a partilha de equipamentos rádio com outros operadores e a ativação de funcionalidades de poupança energética. Através da calibração de um modelo de previsão de consumo energético utilizando dados reais e informações documentadas sobre consumo típico dos equipamentos, foram analisadas as poupanças que podem ser alcançadas com as mudanças propostas. Os resultados mostram que é possível obter uma redução de até 59% nos custos energéticos por site. Nesta dissertação, foi analisado o Custo Total de Propriedade de um operador e de uma empresa detentora de infraestrutura de telecomunicações no mercado português. Por conseguinte, foi desenvolvido um algoritmo em R para determinar o custo de uma torre com base na altura pretendida. A implementação de algoritmos de otimização que se baseiam em técnicas de Inteligência Artificial (IA) é essencial para melhorar a eficiência dos modelos operacionais das redes móveis atuais. Deste modo, analisou-se ainda o impacto médio do recurso a IA no consumo de energia de uma estação base, em que se verificou que através de 2 funcionalidades de poupança energética podem existir poupanças de 156€ por site por ano em torres e 168€ em rooftops. Por fim, foi analisado um caso de investimento em Self-Optimizing Network (SON) e concluiu-se que com apenas 6% de poupança energética, o investimento em técnicas de IA teria retorno em cerca de 2 anos e meio.
- Design of textile antennas for cardiopulmonary applicationsPublication . Cunha, Tiago Alexandre de Oliveira e; Mendes, Carlos; Pinho, PedroVital signs monitoring is a crucial medical application to obtain information about the heath status of a subject. The existing methods existing consist on intrusive and rigid devices, such as the Holter monitors, which are less comfortable for the subject. In order to be able to provide a less intrusive and more comfortable experience for the subject, textile antenna are a suitable solution for this case. Furthermore, textile anten nas can be incorporated in clothes, as they use flexible and light materials. Previous works have been made with textile antennas close to a human body, but mostly radiate in the opposite direction of it. This dissertation presents a study on textile antennas and the possibility to detect vital signs. The process of the textile antenna design will be explained, including the de signs picked and tested to obtain the final textile antenna capable to be in a vicinity of human body and capture vital signs. As well as how the antenna was optimized close to a human body and the method used. The final design was a patch antenna with slots with a resonance frequency of 433.93 MHz and with a superstrate to optimize the an tenna performance close to a human body. The antenna was able to obtain vital signs using a method based on the reflection coefficient phase variation as the respiration and heart cycle occurred.
- Planeamento de viagem em veleiro navegaçãoPublication . Santos, Rui Filipe Guimarães dos; Silva, Paulo Manuel Trigo Cândido da; Trovão, HugoNa concepção de um trajeto marítimo existem diferentes componentes que se devem ter em conta para o seu bom sucesso. O planeamento de um trajeto ponto a ponto não implica apenas a utilização de instrumentos de auxílio mais sofisticados, mas também depende da coordenação e organização da equipa de navegação, um dos fatores mais importantes para uma navegação bem-sucedida. Este trabalho tem como objetivo a criação de um sistema, designado por Travelling Planner, capaz de promover e suportar o trabalho colaborativo para um melhor planeamento de todos os processos inerentes à navegação marítima em diferentes plataformas web e móvel. Estes processos estão na sua maioria relacionados com fatores externos à embarcação, como a previsão meteorológica, a definição de pontos geográficos que constituem o trajeto, a definição da distância mínima e máxima da linha costeira, as ajudas à navegação (e.g., docas, faróis), comparação de trajetos já existentes ou rotas preferidas por outros navegadores, a sumarização do trajeto o definido e a criação de rotas semi-automáticas. A realização deste projeto levou a que fossem executados testes com aproximação ao que seria uma utilização considerada normal, de gestão de um trajeto desde o início do seu planeamento ao fim. Concluindo que com a utilização de diferentes tecnologias e informações públicas, como Open Street Map, Open Sea Map, MapLibre é possível realizar sistemas com funcionalidades de manipulação e extração de informação vetorial para criação de sistemas de informação geográfica.
- Mobile platform for selling used items with recommendations and an image-based system for recognizing similar itemsPublication . Távora, Miguel; Jesus, Rui; Marques, GonçaloThe purpose of this report is to describe the whole process that led to the final product obtained, which is the e-commerce mobile application and the recommendation system of products based on images. The methodologies used in this project are divided into two important key factors. The first one is the impact of software architecture and how it can be applied to explore components like infrastructure needed, technologies, and programming languages. The second one is the importance of the recommendation system in an e-commerce application and how it might en hance the user experience of the application. The software architecture has the objective of creating the whole system from the ground up, utilizing the software engineering principles. The main objective of software engineering should be to develop a system as simple as possible with minimal entropy possible. To achieve this, it is necessary to create the software architecture to use it as a guideline to build the project and also to expand it in the future. This was achieved through the requirement specification, logical and detailed architecture using a top down approach.Despite the time-consuming nature of creating the architecture and its diagrams, in the long run, it helps to reduce the entropy of the system facilitating the development of good quality code with a good organization. The recommendations system proposed is based on searching for similar images. The user uploads one or more images, then the system will use deep learning using a multi class classifier and a distance metric. To evaluate the quality of the system several tests were conducted, obtaining a MAP of 0.897. However, this methodology has some key limitations related to the images themselves. For example, if a user wants a red item but the image has poor illumination or the photo is taken at a bad angle, it might affect the search results.
- Content Management System with Identity Verification and Non-Repudiation SupportPublication . Policarpo, Márcia Krus; Trigo, PauloEvery historical event is perceived in different ways by different people and each person involved has their own testimony of what happened. Traditional social networks and wiki platforms lack the space for personal testimonies. Wikis limit the ability to showcase diverse viewpoints by offering a single-page unified vision. Social networks, characterized by anonymity and ephemeral content, often prioritize the latest trends over preserving individual narratives. To bridge the gap, a knowledge-sharing platform, named ADDA, was developed to commemorate the historical significance of Portugal’s 25th of April. It stands as a testament to the memories of those who experienced this moment in history. The platform is currently publicly available with a subset of its planned features, with ongoing plans for gradual expansion. The ADDA platform aims to balance pseudonymity and identity verification, fostering trust through verification while allowing users the freedom to openly share stories under a pseudonym. This blend promotes responsible sharing, linking accountability with expressive freedom. Digital signatures have been incorporated into the platform to safeguard the authenticity and integrity of each shared testimony (and all of its versions). These signatures serve as a seal of trust, ensuring that the contributions are tamperproof and verifiable. Performance tests were conducted to compare encryption methods, allowing the estab lishment of criteria and the identification of differences for an informed decision on the method to be used. Additionally, usability tests yielded positive feedback and valuable user suggestions. All of this contributes to the ongoing platform enhancements and reinforces the sense of need for a social interaction platform with these characteristics.
- A safety perspective for soft mobility in the ITS ecosystemPublication . Gonçalves, Mafalda Sofia Compadrinho; Datia, Nuno Miguel SoaresAll road users have one concern in common, regardless of the mode of transport they use — reaching their destination safely. Users of soft mobility have a high fatality rate, meaning that every time they choose to walk or drive a bicycle, for example, they are ultimately putting their lives at risk. It was found that these entities’ safety must be protected, and this motivation is the major reason for designing a tool that will help decrease the number of road accidents and their severity, along with a desire to reduce carbon dioxide emissions by using fuel-efficient vehicles. In this project, a technological system based on information provided by ITS messages, as well as GPS location data, was implemented for users of soft mobility so they can be aware of dangers that show up in their travel path, as well as promoting the use of soft mobility as an al ternative mode of transportation. The system has a UI that allows users to receive information without disturbing or diverting their attention from the path ahead for a long time, showing notifications for potentially dangerous road events or other road entities approaching. There is equipment that allows the exchange of messages related to vehicles, roads and traffic, which operates over ITS-G5. To ensure soft mobility users are integrated into the ITS ecosystem, a hybrid network environment that allows com munications between cellular and ITS-G5 networks was established. A prototype of this system was designed and tested on multiple use cases, in a real and controlled environment, addressing critical scenarios for the user’s safety. According to the results, with the latency involved in E2E communications, this network can support several urban scenarios, where soft mobility users and connected cars share the road. However, in order to guarantee soft mobility user’s safety, further testing should be done accounting for the processing time of the application and user’s time-to-react.
- Driver Profile ClassificationPublication . Loureiro, Luís Miguel Pinto; Lourenço, André Ribeiro; Ferreira, Artur JorgeNowadays, we have billions of vehicles around the world. In most countries, insurance against civil liability for vehicles is mandatory. Usually, insurance companies define vehicle insurance rates according to static known time-invariant variables or with a slight adjustment over time. The insurance rates calculation relies mostly on factors such as the age of the driver, the number of years one holds a driving license, and the driving history. These variables are not in close connection with the everyday behavior of the driver on the road, thus being not fair for young drivers, for instance. In this thesis, we follow a pay-as-you-behave approach, in which we describe the research and development work to devise a driver profile classification solution, based on driver behavior data. From data records with the trips from different drivers, we build a dataset, targeting a driver profile classification task. We explore and evaluate the use of machine learning techniques over the dataset, starting with an unsupervised approach and moving to a supervised one, where we save the final model. We also develop a REST API to serve as proof of concept to classify drivers and fleets, that uses the best model to determine the driver profile. The experimental results show that unsupervised machine learning techniques can cluster and identify distinct trip profiles, and supervised learning techniques are capa ble of classifying different trips based on the profiles found, which are then used to assign the driver profile. Many drivers revealed to have inconsistent driver behavior over time. The majority of the drivers has a non-aggressive driving style, some are aggressive, and a few are risky style drivers.
- Robust cardiopulmonary antenna design for any body mass indexPublication . Cardoso, João Filipe Ferreira da Silva Tenente; Pinho, Pedro; Albuquerque, DanielWith the advancement of wireless diagnostic and treatment technologies, antennas implanted inside and outside the human body are now widely used. The use of antennas on the body, along with other technologies, presents itself as an innovative method for detecting and monitoring vital signs. These antennas can be placed in direct contact with the human body or on clothing making vital signs monitoring with these methods more comfortable and less invasive when compared to traditional methods, allowing home monitoring of elderly patients or high risk workers such as police, firefighters and military personnel. A gap was found in the literature regarding the development of a standard antenna model that can be used by anyone, regardless of their physiog nomy. In this dissertation, a robust high bandwidth antenna was developed to operate in the ISM frequency band, namely at 2.45 GHz, capable of monitoring vital signs in any subject. In order to increase its bandwidth, parasitic patches placed on top of the radiating element were used to obtain resonances adjacent to the center frequency, thus increasing the bandwidth of the antenna. After the antenna design, the antenna was simulated, built, and the vital signs monitoring of 5 subjects with different body physiognomies was performed. The effectiveness of the chosen design was proven, being able to monitor the respiratory signal in all test subjects.