ISEL - Engenharia Electrónica, Telecomunicações e Computadores
URI permanente desta comunidade:
Navegar
Percorrer ISEL - Engenharia Electrónica, Telecomunicações e Computadores por Domínios Científicos e Tecnológicos (FOS) "Engenharia e Tecnologia::Engenharia Eletrotécnica, Eletrónica e Informática"
A mostrar 1 - 10 de 45
Resultados por página
Opções de ordenação
- 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.
- Aplicações de modelos de linguagem de grande escala na cibersegurançaPublication . Conceição, Tiago Miguel Pestana; Cruz, Nuno Miguel MachadoA crescente complexidade e sofisticação das ameaças no ciberespaço têm impulsionado a procura por soluções inovadoras e eficientes no campo da cibersegurança. Neste contexto, conduziu-se uma investigação com o objetivo de avaliar a viabilidade de Large Language Models (LLMs) no que concerne à automatização da geração de código e configurações no âmbito da cibersegurança. A investigação centrou-se em mecanismos de ciberdefesa e aplicações de educação em cibersegurança, com particular ênfase em soluções de geração de honeypots, malware e exercícios de Capture The Flag (CTF). Foram avaliados sete modelos, incluindo o GPT-4, Gemini Pro e Claude Opus 3. A metodologia de avaliação assentou no desenvolvimento de dois mecanismos de avaliação, sendo o primeiro um novo benchmark Cybersecurity Language Understading (CSLU), baseado no Massive Multitask Language Understanding, constituído por questões de escolha múltipla sobre diversos domínios do conhecimento. As prompts foram concebidas com o intuito de avaliar o estado de conhecimento de cada modelo relativamente aos tópicos supracitados. O segundo mecanismo consistiu na avaliação da consistência, criatividade e adaptabilidade dos modelos referente à geração de artefactos. Os resultados evidenciaram uma notória proeminência referente ao tópico de malware, com quatro destes a alcançarem a pontuação máxima. Por outro lado, o desempenho na tarefa de CTF revelou uma maior variação de resultados. De um modo geral, os modelos GPT- 4, Gemini Pro e Claude 3 Opus demonstraram resultados consistentemente superiores entre os modelos estudados. Num segundo momento, pretendeu-se desenvolver uma ferramenta baseada na web, com o objetivo de fornecer uma prova de conceito dos estudos anterior realizados. A referida ferramenta, recorrendo aos melhores LLMs estudados, permite ao utilizador criar e lançar automaticamente serviços de segurança, como os mencionados honeypots ou exercícios de CTF. De uma perspetiva global, estas descobertas sugerem que a aplicação de LLMs em atividades de cibersegurança pode ser altamente vantajosa.
- Assessing machine learning techniques for intrusion detection in cyber-physical systemsPublication . Santos, Vinicius F.; Albuquerque, Célio; Passos, Diego; Ereno Quincozes, Silvio; Mossé, DanielCyber-physical systems (CPS) are vital to key infrastructures such as Smart Grids and water treatment, and are increasingly vulnerable to a broad spectrum of evolving attacks. Whereas traditional security mechanisms, such as encryption and firewalls, are often inadequate for CPS architectures, the implementation of Intrusion Detection Systems (IDS) tailored for CPS has become an essential strategy for securing them. In this context, it is worth noting the difference between traditional offline Machine Learning (ML) techniques and understanding how they perform under different IDS applications. To answer these questions, this article presents a novel comparison of five offline and three online ML algorithms for intrusion detection using seven CPS-specific datasets, revealing that offline ML is superior when attack signatures are present without time constraints, while online techniques offer a quicker response to new attacks. The findings provide a pathway for enhancing CPS security through a balanced and effective combination of ML techniques.
- Assistente digital baseado em inteligência artificial para PCPublication . Goulão, António Duarte; Leite, Nuno Miguel da Costa de Sousa; Ferreira, Artur JorgeNos últimos anos, os assistentes digitais tornaram-se cada vez mais populares como meio de interação entre utilizadores e sistemas computacionais. No entanto, a maioria das soluções existentes é proprietária e fortemente integrada em ecossistemas fechados, limitando a flexibilidade e transparência. Esta tese propõe o desenvolvimento de um Assistente Digital (AD) para ambiente desktop Windows, modular, extensível e baseado em tecnologias de acesso aberto. O sistema integra reconhecimento automático de fala (ASR), síntese de fala (TTS), processamento de linguagem natural (PLN) e uma interface gráfica interativa. A arquitetura modular permite substituir ou expandir funcionalidades sem comprometer o núcleo do sistema. Um modelo de linguagem em larga escala (LLM) é utilizado para interpretar comandos em linguagem natural, garantindo flexibilidade na compreensão de instruções. Foram implementadas funcionalidades como execução de comandos locais e integração com serviços externos (Google Calendar e Gmail). Todos os comandos foram avaliados com LLM de diferentes dimensões. Os resultados mostraram que o desempenho está diretamente ligado ao modelo utilizado: modelos menores apresentaram falhas ocasionais, enquanto os de maior escala garantiram elevada precisão e consistência. Em todos os casos, os tempos médios de resposta mantiveram-se baixos, na ordem dos décimos de segundo. Para avaliar a usabilidade, foi aplicado um questionário baseado na métrica SUS a 15 utilizadores, com resultados muito positivos (pontuação média de 90.33 em 100). Os participantes mostraram facilidade na execução das tarefas e sugeriram melhorias relevantes. A solução confirma a viabilidade de um AD modular, expansível e open source para desktop. O trabalho constitui uma base sólida para futuras evoluções, permitindo a integração de novos módulos e adoção de diferentes LLM, representando um passo relevante no desenvolvimento de assistentes digitais mais abertos e adaptáveis.
- An autonomic parallel strategy for exhaustive search tree algorithms on shared or heterogeneous systemsPublication . Gonçalves de Oliveira Passos, Fernanda; Rebello, Vinod E. F.Backtracking branch-and-prune (BP) algorithms and their variants are exhaustive search tree techniques widely employed to solve optimization problems in many scientific areas. However, they characteristically often demand significant amounts of computing power for problem sizes representative of real-world scenarios. Given that their search domains can often be partitioned, these algorithms are frequently designed to execute in parallel by harnessing distributed computing systems. However, to achieve efficient parallel execution times, an effective strategy is required to balance the nonuniform partition workloads across the available resources. Furthermore, with the increasing integration of servers with heterogeneous resources and the adoption of resource sharing, balancing workloads is becoming complex. This paper proposes a strategy to execute parallel BP algorithms more efficiently on even shared or heterogeneous distributed systems. The approach integrates a self-adjusting dynamic partitioning method in the BP algorithm with a dynamic scheduler, provided by an application middleware, which manages the parallel execution while addressing any issues of imbalance. Empirical results indicate better scalability with efficiencies above 90% for instances of an application case study for the discretizable molecular distance geometry problem (DMDGP). Improvements of up to 38% were obtained in execution speed-ups compared to a more traditional parallel BP implementation for DMDGP.
- Bitcoin Anomaly Detection (BAD) - use of machine learning for fraudulent transaction detectionPublication . Gomes, Nuno Gonçalo Rodrigues Cabral; Ferreira, Artur JorgeAbstract The increasing adoption of cryptocurrencies, especially Bitcoin, has significantly altered the financial landscape, enabling decentralised and pseudonymous transactions. While these characteristics foster innovation, they also present serious challenges in detecting fraudulent behaviour, including money laundering and investment scams. This dissertation introduces Bitcoin Anomaly Detection (BAD), a hybrid machine learning framework designed to detect anomalous transactions on the Bitcoin blockchain. The methodology integrates supervised and unsupervised learning techniques, applied to the Elliptic dataset comprising over 200,000 transactions with temporal and graph-based features. Feature engineering, class imbalance handling (e.g., SMOTE + ENN), and dimensionality reduction (PCA, UMAP) are employed. Several models are evaluated, including XGBoost, RF, and GNN, achieving up to 97.5% accuracy, 96.2% recall, and a false positive rate below 4.5%. Graph analysis revealed that illicit transactions tend to form sparsely connected “sink” nodes—receiving many inputs (high in-degree) but sending no outputs (zero out-degree)—a pattern typical of laundering. Semi-supervised learning and association rule mining were used to label unknown data and enhance classification reliability. The final pipeline combines feature selection, class instance sampling, and a hybrid semisupervised learning approach—bridging supervised and unsupervised methods—to classify transactions with high accuracy and robustness. Key challenges addressed include the scarcity of labelled data, severe class imbalance, and the evolving nature of fraud techniques. XAI components were incorporated to ensure interpretability and compliance with regulatory frameworks such as Markets in Crypto-Assets (MiCA) and Financial Action Task Force (FATF). The findings from our experimental evaluation demonstrate the viability of adaptive, interpretable AI solutions for safeguarding decentralised financial ecosystems and supporting efforts in Anti-Money Laundering (AML) and law enforcement.
- Centralized ledger system for document and process certificationPublication . Bartolomeu, Nuno António Oliveira; Leite, Nuno Miguel da Costa de Sousa; Pereira, João Miguel de Carvalho da ConceiçãoAbstract Organizations increasingly require secure document management with integrity guarantees beyond traditional audit logs, particularly in regulated industries where external accountability is critical. While blockchain technologies provide strong tamper-detection, they present significant enterprise adoption challenges including cost volatility, low throughput, and unpredictable operational expenses. This thesis proposes a Centralized Ledger System (CLS) that provides blockchain-inspired integrity verification through self-hosted architecture without external dependencies. The system implements a three-phase entry lifecycle, signature collection and verification, supporting multi-party transactions with asynchronous workflows. A multi-ledger architecture enables organizational segregation of business domains while maintaining referential integrity. Key contributions include automated receipt generation for independent verification, selective payload erasure preserving cryptographic validation, entry linking for audit simplification, and integration of security services with two-factor authentication and key management. The modular design enables flexible deployment while maintaining cryptographic guarantees equivalent to blockchain systems. The solution addresses the gap between traditional audit systems and distributed ledgers by providing cost-predictable, vendor-independent functionality that integrates into existing workflows without specialized blockchain expertise.
- Cognitive radio with machine learning to increase spectral efficiency in indoor application on the 2.5 GHz bandPublication . Soares, Marilson Duarte; Passos, Diego; Castellanos, Pedro Vladimir GonzalezDue to the propagation characteristics in the 2.5 GHz band, the signal is significantly degraded by building entry loss (BEL), making coverage in indoor environments in some cases non-existent. Signal degradation inside buildings is a challenge for planning engineers, but it can be seen as a spectrum usage opportunity for a cognitive radio communication system. This work presents a methodology based on statistical modeling of data collected by a spectrum analyzer and the application of machine learning (ML) to leverage the use of those opportunities by autonomous and decentralized cognitive radios (CRs), independent of any mobile operator or external database. The proposed design targets using as few narrowband spectrum sensors as possible in order to reduce the cost of the CRs and sensing time, as well as improving energy efficiency. Those characteristics make our design especially interesting for internet of things (IoT) applications or low-cost sensor networks that may use idle mobile spectrum with high reliability and good recall.
- Cross-device platform for collaborative and immersive experiences in mixed realityPublication . Lucas, Letícia; Costa, Carla; Jorge, Pedro MendesThis work aims to develop a platform for managing Augmented Reality (AR) and Virtual Reality (VR) devices, enabling multiple users to engage in collaborative and immersive experiences occurring both in the physical and digital world. The proposed solution addresses challenges in synchronization, data transmission, and real-time rendering, so that the multiple participants - either co-located or geographically dispersed - can share a unified, high-fidelity experience. This research intends to advance mixed reality technologies for education, entertainment, and business purposes. To demonstrate this cross-device approach, an application using the MagicLeap2 and MetaQuest3 devices is implemented. This application allows various users to collaboratively interact within the same mixed reality environment. Unity3D is the core framework of this system, and OpenXR is the layer that ensures hardware compatibility.
- Cross-device platform for collaborative and immersive experiences in mixed realityPublication . Lucas, Letícia; Costa, Carla; Jorge, Pedro MendesThis work aims to develop a platform for managing Augmented Reality (AR) and Virtual Reality (VR) devices, enabling multiple users to engage in collaborative and immersive experiences occurring both in the physical and digital world. The proposed solution addresses challenges in synchronization, data transmission, and real-time rendering, so that the multiple participants - either co-located or geographically dispersed - can share a unified, high-fidelity experience. This research intends to advance mixed reality technologies for education, entertainment, and business purposes. To demonstrate this cross-device approach, an application using the MagicLeap2 and MetaQuest3 devices is implemented. This application allows various users to collaboratively interact within the same mixed reality environment. Unity3D is the core framework of this system, and OpenXR is the layer that ensures hardware compatibility.