Research

In my research, I explore how artificial intelligence can be used to improve cybersecurity and decentralize architectures. I am particularly interested in federated learning, a type of machine learning that allows multiple devices to train models on data while keeping that data private. I believe that this approach has great potential for improving both security and privacy in many applications.

In parallel, I study the application of Brain-Computer Interfaces (BCIs) to control remote devices. This work has the potential to provide people with disabilities greater independence and control over their environment.

  • Artificial Intelligence (AI)
  • Cybersecurity
  • Decentralized architectures
  • Federated Learning
  • Adversarial attacks
  • Brain-Computer Interfaces

Laboratory Colleagues

Manuel Gil Pérez

Manuel Gil Pérez

Ph.D. - Associate Professor

Follow
Alberto Huertas Celdrán

Alberto Huertas Celdrán

Ph.D. - Postdoctoral Researcher Associated

Follow
Gregorio Martínez Pérez

Gregorio Martínez Pérez

Ph.D. - Professor (Full)

Follow
Gérôme Bovet

Gérôme Bovet

Head of Data Science chez armasuisse W+T

Follow
Sergio López Bernal

Sergio López Bernal

Ph.D.

Follow
Felix Gómez Mármol

Felix Gómez Mármol

Ph.D. - Professor (Full)

Follow
Pedro Miguel Sánchez Sánchez

Pedro Miguel Sánchez Sánchez

Ph.D. Student

Follow

Great team members!

Currently, I am part of the CyberDataLab (Cybersecurity and Data Science Lab) at the University of Murcia. This lab «encompasses a group of skilled and motivated scientists and researchers with a strong commitment for quality research and developments around the two main topics of the lab: cybersecurity and data science. It is worth mentioning the value created in the intersection of both fields, with a strong projection onto international R&D programs such as Horizon Europe, European Defence Agency R&T or National Science Foundation, amongst others».

Research Projects

  • DEFENDIS: Decentralized Federated Learning for IOT Device Identification and Security

    Federal Office for Defence Procurement armasuisse, 04/02/2023 – 30/11/2023

    DEFENDIS: DEcentralized FEderated learNing for IoT Device Identification and Security project aims to develop a framework designed to uniquely identify each device deployed in an IoT platform in a distributed and robust manner, solving possible security threats based on device impersonation or malicious deployment.

    The proposed framework is based on hardware device fingerprinting and the fully distributed generation of ML/DL models to identify these devices as well as possible malicious elements affecting the identification process robustness. Besides, the monitoring of the processes running on the device is also considered as a contextual data source to be employed during environment securitization. The main objectives of the platform are:

    • To provide a solution to uniquely identify each of the sensors of a crowdsensing or Industrial IoT (IIoT) platform in a reliable manner, strongly solving possible sensor impersonation security threats. In this sense, the solution needs to monitor also contextual information such as running processes, temperature, or CPU load in order to adjust the parameters of the generated fingerprint according to its context. The privacy-preserving management and exchange of device fingerprints and models is based on Federated Learning (FL).
    • To apply adversarial attacks against the solution and identify their proper countermeasures, improving its resilience against possible malicious actors taking part in the federation. These attacks will target both the fingerprint generation and the FL model training and deployment process, so the complete solution lifecycle is secured.
    • To develop a fully Decentralized FL (DFL) framework for ML/DL model generation, enabling model training and distributing the fingerprints across different stakeholders without the requirement of sharing sensitive information or having a central entity managing the aggregation of the models, reducing the bottleneck and attack surface of having a centralized server.
    • To analyze the main trust and robustness metrics related to the FL model generation process and integrate them into the framework developed in the previous point. Some metrics considered are robustness, privacy, fairness, accountability, and explainability.
  • EU-GUARDIAN: European Framework and Proofs-of-concept for the Intelligent Automation of Cyber Defence Incident Management

    European Defence Fund, 13.454.545.33€, 01/12/2022 – 01/11/2025

    EU-GUARDIAN aims at creating a cutting-edge, accurate and reliable AI-based solution that operates and automates larger parts of incident management and cyber defence processes. Focusing on the ability to detect, mitigate and respond to security challenges semi-automatically or automatically; support analysts and decision-makers at all levels; and contribute to enhancing cyber situational awareness, military infrastructure resilience and protection against advanced cyber threats.

    On these grounds, EU-GUARDIAN will make progress on understanding the relevant actors and the threat environment where AI-based cyber defence systems will interact with; creating AI-based techniques for detecting and understanding adversarial activity, as well as for building knowledge about own protected ICT systems; shaping AI-based information collection and storage systems that can dynamically adapt their strategy to the situation perceived; and building AI-based decision systems which are risk and impact aware. All components will follow the key requirements of human agency and oversight; technical robustness; privacy; algorithmic transparency; diversity and accountability; and proof-of-concept feasibility analysis will support each component.

    The multidisciplinary and highly specialised Consortium will present the results of EU-GUARDIAN, which will facilitate the tedious task of analysing large amounts of data; will improve cyber operational capability; drive a reduction in costs; and above all, they will contribute to EU cyber defence posture and to the laying of the foundations for prompting the EU autonomy in development and capacitation of AI-based resources.

  • VALKYRIES: Harmonization and Pre-Standardization of and Tactical Coordinated procedures for First Aid Vehicles deployment on European multi-victim Disasters

    European Commission, H2020-ICT, 101020676, 5.995.757,50€, 01/10/2021 – 30/09/2023

    Objective

    H2020-VALKYRIES will develop, integrate and demonstrate capabilities for enablingimmediate and coordinated emergency response including search and rescue, security and health, in scenarios of natural/provoked catastrophes with multiple victims, with special application in cases in which several regions or countries are affected and hence greater interoperability being required. H2020-VALKYRIES will propose both design and development of a modular, interoperable, scalable and secure platform, which will allow the integration between legacy solutions and new technologies. The platform will be able to deploy services and dynamically adapt its behaviour, as the emergency requires it. A series of use cases and demonstrators will be developed placing an emphasis on cross-frontier and cross-sectorial BLOS (Beyond Line of Sight) scenarios, where the usual communications infrastructure could have been damaged, and emergency response teams are deployed without an accurate view of the operation environment.

    Participants

    Indra (Spain, Coordinator), Servicio Madrileño de Salud (Spain), Tassica Emergency Training & Research (Spain), ISEM – Inštitút pre medzinárodnú bezpečnosť at krízové riadenie, n. o (Slovakia), Universidad de Murcia (Spain), Scuola Superiore Sant’Anna (Italy), Blockchain2050 BV (Netherlands), Institut Po Otbrana (Bulgaria), Bulgarian Red Cross (Bulgaria), Kentro Meleton Asfaleias (Greece), Hospital do Espírito Santo de Évora (Portugal), Aratos Ntot Net LTD (Greece), University of South-Eastern Norway (Norway), Agenzia Regionale Emergenza Urgenza (Italy), Hellenic Rescue Team (Greece), Novotec Consultores (Spain), Particle Summary (Portugal)

  • CyberBrain: Cybersecurity in BCI for Advanced Driver Assistance

    Bitbrain Technologies S.L.

    Objective

    Cybersecurity in BCI has barely been studied in the literature, counting only with few and limited works implementing proofs of concept in marginal scenarios. Based on that, the main objective of CyberBrain is to design and implement a framework able to detect cyberattacks affecting the BCI lifecycle while using Bitbrain products. After analyzing the BCI and Bitbrain vulnerabilities and proposing a list of countermeasures, CyberBrain will design and deploy a set of cyberattacks targeting three challenging use cases that integrates Bitbrain products with an advanced driver assistance scenario. The cyberattacks impact will be measured by the framework through a set of metrics defined during the project and provided as outcome of CyberBrain. Finally, the previous contributions and additional private and public documents, paper, software, and videos will benefit Bitbrain and the BCI community, respectively.

  • Framework for distraction detection in driving scenarios using Brain-Computer Interfaces

    Master's Thesis at the University of Murcia

  • Framework para detección de ondas P300 y ciberataques basados en ruido en Interfaces Cerebro-Máquina

    Final Degree Project at the University of Murcia

  • COnVIDa - COVID19 data monitoring in Spain

    COnVIDa is a tool developed by the Cybersecurity and Data Science Laboratory (CyberDataLab) of the University of Murcia that easily collects data related to the COVID19 pandemic from different data sources in the context of Spain