DIstributed Smart Communications with Verifiable EneRgy-optimal Yields

DISCOVERY is a coordinated research project conducted by teams from three different Spanish universities (UVigo, UPC, and UC3M) with complementary research skills. The project combines our knowledge in the areas of distributed learning, energy-efficiency, security, data privacy, communication protocols, and advanced services to face improvements in distributed communication protocols and systems with privacy and security protection mechanisms, jointly designed so that the digital carbon footprint (DCF) of the entire system can be characterized, measured, and optimized according to the users’ requirements.

The research goals in DISCOVERY will contribute to the realization of more efficient, secure, and sustainable distributed digital infrastructures currently used for data-driven smart services powered by statistical learning models.

Download brochure (PDF)
Evaluation of the level of privacy in FL

The aim is to define privacy metrics in FL that establish how much information an active adversary can obtain a priori (before running a given number of interactions of the learning algorithm) and a posteriori (after running that number of interactions). Real-time privacy loss estimation methods will be developed.

Design of secure aggregation techniques in (D)FL to prevent attacks against the aggregator(s) or the learning network computation nodes

Homomorphic encryption, secure multiparty computation and differential privacy techniques will be applied. Much of these techniques provide post-quantum resistance in a natural way, since they are either based on hard problems over lattices, or provide theoretical security under certain assumptions of non-collusion.

Analysis of threat models in FL

The aim is to identify realistic threats in FL environments where active adversaries try to infer as much information as possible during the learning process.

Design of attacks with active adversaries in FL

We intend to design membership inference attacks (to know if a subject belongs to a training group) and property inference attacks (to validate if a record in the database satisfies a property).

Evaluation of the level of privacy in FL

The aim is to define privacy metrics in FL that establish how much information an active adversary can obtain a priori (before running a given number of interactions of the learning algorithm) and a posteriori (after running that number of interactions). Real-time privacy loss estimation methods will be developed.

Design of secure aggregation techniques in (D)FL to prevent attacks against the aggregator(s) or the learning network computation nodes

Homomorphic encryption, secure multiparty computation and differential privacy techniques will be applied. Much of these techniques provide post-quantum resistance in a natural way, since they are either based on hard problems over lattices, or provide theoretical security under certain assumptions of non-collusion.

Design of authentication and accountability algorithms in (D)FL

This involves integrating Blockchain, authentication and direct model sharing technologies into a decentralized learning architecture to ensure traceability of learning processes.

Improving the efficiency and robustness of (D)FL algorithms to deal with adverse situations such as failures, interruption or delay in communications, and statistical inhomogeneity in the data

Improving the efficiency and robustness of (D)FL algorithms to deal with adverse situations such as failures, interruption or delay in communications, and statistical inhomogeneity in the data. The aim is to optimize the cost of communication and computation while maintaining the models in an adequate degree of updating depending on their availability and the conditions of connection and delay in the network.

Create a toolbox to experiment countermeasures in the private, secure and robust operation of (D)FL systems

This software will be used to generate a prototype (use case) in the finance field to detect fraudulent transactions.

O1

Analysis of threat models in FL

The aim is to identify realistic threats in FL environments where active adversaries try to infer as much information as possible during the learning process.

O2

Design of attacks with active adversaries in FL

We intend to design membership inference attacks (to know if a subject belongs to a training group) and property inference attacks (to validate if a record in the database satisfies a property).

O3

Evaluation of the level of privacy in FL

The aim is to define privacy metrics in FL that establish how much information an active adversary can obtain a priori (before running a given number of interactions of the learning algorithm) and a posteriori (after running that number of interactions). Real-time privacy loss estimation methods will be developed.

O4

Design of secure aggregation techniques in (D)FL to prevent attacks against the aggregator(s) or the learning network computation nodes

Homomorphic encryption, secure multiparty computation and differential privacy techniques will be applied. Much of these techniques provide post-quantum resistance in a natural way, since they are either based on hard problems over lattices, or provide theoretical security under certain assumptions of non-collusion.

O5

Design of authentication and accountability algorithms in (D)FL

This involves integrating Blockchain, authentication and direct model sharing technologies into a decentralized learning architecture to ensure traceability of learning processes.

O6

Improving the efficiency and robustness of (D)FL algorithms to deal with adverse situations such as failures, interruption or delay in communications, and statistical inhomogeneity in the data

Improving the efficiency and robustness of (D)FL algorithms to deal with adverse situations such as failures, interruption or delay in communications, and statistical inhomogeneity in the data. The aim is to optimize the cost of communication and computation while maintaining the models in an adequate degree of updating depending on their availability and the conditions of connection and delay in the network.

O7

Create a toolbox to experiment countermeasures in the private, secure and robust operation of (D)FL systems

This software will be used to generate a prototype (use case) in the finance field to detect fraudulent transactions.

ACTIVITIES

Partners

Atlanttic
logo-uc3m

Rebeca P. Díaz Redondo

Principal Investigator

Fernando Pérez González

Principal Investigator

Manuel Fernández Veiga

Researcher

Ana Fernández Vilas

Researcher

Alberto Pedrouzo Ulloa

Researcher

Pedro Comesaña Alfaro

Researcher

David Vázquez Padín

Researcher

Results

Conference Papers

  1. M. Rashed, C. García-Rubio, C. Campo, “Studying MQTT malicious traffic: methods and challenges,” en Actas de las X Jornadas Nacionales de Investigación en Ciberseguridad, 2025, pp. 150–156.
  2. M. Moure-Garrido, C. Campo, y C. García-Rubio, “Fingerprinting Encrypted DNS: Exploiting Metadata Leakage in DNS over QUIC,” en Proceedings of the 21st International Symposium on Performance Evaluation of Wireless Ad Hoc, Sensor and Ubiquitous Networks (PE-WASUN), Barcelona, España, oct. 2025. DOI: 10.1109/MSWiM67937.2025.11309217
  3. A. Jimenez-Berenguel, C. Campo, y M. Moure-Garrido, “Enhancing Privacy in DNS Communications with Energy-Aware Methodologies,” en Proceedings of the 27th International Conference on Modeling, Analysis and Simulation of Wireless and Mobile Systems (MSWiM) PhD Forum, Barcelona, España, oct. 2025. 10.1109/MSWiM67937.2025.11308995
  4. J. L. Mejía-Acuña, D. Díaz-Sánchez, F. Almenárez-Mendoza, C. Campo-Vázquez, y C. García-Rubio, “Evaluating Verifiable Data Structures for Trustworthy Environmental and Consumption Credentials,” en Proceedings of the 17th International Conference on Ubiquitous Computing and Ambient Intelligence (UCAmI), nov. 2025.
  5. M. Moure-Garrido, S. Anillo-Baeza, C. García-Rubio, y C. Campo, “DoH: A Double-Edged Sword for Privacy? Unmasking Hidden Traffic Patterns with Machine Learning,” en Proceedings of the 17th International Conference on Ubiquitous Computing and Ambient Intelligence (UCAmI), nov. 2025.
  6. J. Gento Suela, J. Blanco-Romero, F. Almenárez Mendoza, D. Díaz Sánchez, «Implementing and Evaluating Post-Quantum DNSSEC in CoreDNS», en Proceedings of the 21st International Symposium on Performance Evaluation of Wireless Ad Hoc, Sensor and Ubiquitous Networks (PE-WASUN), Barcelona, España, oct. 2025. 10.1109/MSWiM67937.2025.11308970
  7. J. Blanco-Romero, Y. García Niño, F. Almenares Mendoza, D. Díaz-Sánchez, C. García Rubio, C. Campo, «Post-Quantum Entropy as a Service for Embedded Systems», en Proceedings of the 17th International Conference on Ubiquitous Computing and Ambient Intelligence (UCAmI), nov. 2025.
  8. Alberto Bazán Guillén, Pablo A. Barbecho Bautista, Mónica Aguilar Igartua, «RUTGe: Realistic Urban Traffic Generator for urban environments using Deep Reinforcement Learning and SUMO simulator», 11th International Conference on Vehicle Technology and Intelligent Transport Systems (VEHITS 2025), pp 557-564, DOI:10.5220/0013375000003941 Porto, Portugal. April 2nd – 4th, 2025. CORE-C.
  9. Ana Laura Pérez Méndez, René Monteagudo Gordillo, Carlos Alberto Bazan Prieto, Alberto Bazán Guillén, Rafael E. Bello Pérez and Mónica Aguilar Igartua, » Artificial Intelligence Methods for Anomaly Detection in Energy Consumption», 21st IEEE International Symposium on Performance Evaluation of Wireless Ad Hoc, Sensor, and Ubiquitous Networks (PE-WASUN 2025), Barcelona, Spain. October 26th – 31st, 2025. CORE-C. 10.1109/MSWiM67937.2025.11308730
  10. Alberto Bazán-Guillén, Pablo Barbecho-Bautista, Mónica Aguilar Igartua, Francesca Cuomo, «Comparing Optimal and Adaptive EV Charging in Smart Cities: MILP vs. Reinforcement Learning», 21st IEEE International Symposium on Performance Evaluation of Wireless Ad Hoc, Sensor, and Ubiquitous Networks (PE-WASUN 2025), Barcelona, Spain. October 26th – 31st, 2025. CORE-C. 10.1109/MSWiM67937.2025.11308997
  11. Agneev Guin, Alberto Bazán Guillén, Prashanth Kannan and Mónica Aguilar Igartua, «Simulation under Stress: A Comparative Benchmarking of Large-Scale Traffic Simulators», 21st IEEE International Symposium on Performance Evaluation of Wireless Ad Hoc, Sensor, and Ubiquitous Networks (PE-WASUN 2025), Barcelona, Spain. October 26th – 31st, 2025. CORE-C. 10.1109/MSWiM67937.2025.11309008
  12. Juan Pablo Pérez Vargas, Jorge Geovanny Zhangallimbay Coraizaca, Alberto Bazán-Guillén, Pablo Barbecho Bautista, Mónica Aguilar Igartua, «Gap-Fuzzy Adaptive Signal Control: Enhancing Urban Traffic Efficiency», 21st IEEE International Symposium on Performance Evaluation of Wireless Ad Hoc, Sensor, and Ubiquitous Networks (PE-WASUN 2025), Barcelona, Spain. October 26th – 31st, 2025. CORE-C. 10.1109/MSWiM67937.2025.11308772
  13. Agneev Guin, Mónica Aguilar Igartua, «An AI-based Intelligent Vehicle Routing approach for Large-scale Fleet optimization», 27th IEEE International Conference on Modeling, Analysis and Simulation of Wireless and Mobile Systems. (MSWiM 2025), Barcelona, Spain. October 26th – 31st, 2025. CORE-A. 10.1109/MSWiM67937.2025.11308747
  14. Yaqoob Al-Zuhairi, Aya Maher, Alberto Bazán, Mónica Aguilar Igartua, «Federated Learning-Based Electric Vehicle Energy Consumption Prediction and Charging Station Recommendation», 27th IEEE International Conference on Modeling, Analysis and Simulation of Wireless and Mobile Systems. (MSWiM 2025), Barcelona, Spain. October 26th – 31st, 2025. CORE-A. 10.1109/MSWiM67937.2025.11309129
  15. Argenis Andrade, Leticia Lemus Cárdenas, Juan Pablo Astudillo León, Manuel Eugenio Morocho-Cayamcela and Luis J. de la Cruz Llopis, «Performance evaluation of DQN-Based Reinforcement Learning Congestion Control in Mobile Multi-Hop Wireless Networks», 21st IEEE International Symposium on Performance Evaluation of Wireless Ad Hoc, Sensor, and Ubiquitous Networks (PE-WASUN 2025), Barcelona, Spain. October 26th – 31st, 2025. CORE-C. 10.1109/MSWiM67937.2025.11308763
  16. Zari, O., Xu, C., Parra-Arnau, J., Ünsal, A, Önen, M. (2024): Link inference attacks in vertical federated graph learning. In: Annual Computer Security Applications Conference (ACSAC), Hawaii, USA, Dec., pp. 761-777. DOI. CORE: A. DOI: 10.1109/ACSAC63791.2024.00068
  17. Xavier Martínez-Luaña, Manuel Fernández-Veiga and Rebeca P. Díaz-Redondo. “Privacy-aware Berrut Approximated Coded Computing applied to Federated Learning”. JNIC’25
  18. D. Cajaraville-Aboy, Ana Fernández-Vilas, R. P. Díaz-Redondo, M. Fernández-Veiga. «Decentralized Orchestration Framework for Distributed AI Deployments under Fluid Computing Environments». PE-WASUN’25. DOI: 10.1109/MSWiM67937.2025.11308919
  19. Álvaro Vázquez, M. Fernández-Veiga, C. Giraldo. «Diffusion-based Solver for CNF Placement on the Cloud-Continuum». PE-WASUN’25, arXiv: 2511.01343 DOI: 10.1109/MSWiM67937.2025.11309048
  20. Alfonso Camblor, M. Fernández-Veiga. «A Zero-Trust Architecture for Private Collective Forecasting». IEEE CSR’26, Lisboa.

Journal Papers

  1. Jiménez-Berenguel, C. Gil, C. García-Rubio, J. Forné, y C. Campo, “DNS Query Forgery: A Client-Side Defense Against Mobile App Traffic Profiling,” IEEE Access, 2025, doi: 10.1109/ACCESS.2025.3633695
  2. J. D. Llano-Miraval, C. Campo, C. García-Rubio, y M. Moure-Garrido, “AI Versus IoT Security: Fingerprinting and Defenses Against TLS Handshake-Based IoT Device Classification,” IEEE Access, vol. 13, pp. 165607–165622, 2025, doi: 10.1109/ACCESS.2025.3611160
  3. M. Rashed, I. T.-D. Viso, y A. I. González-Tablas, “A Comparison of Cyber Intelligence Platforms in the Context of IoT Devices and Smart Homes,” Electronics, vol. 14, no. 22, art. no. 4503, 2025, doi: 10.3390/electronics14224503
  4. D. Díaz-Sánchez, F. Almenárez, C. Campo, C. García-Rubio, y S. Sherratt, “Beyond PKI: A DNSSEC Delegation Approach for Scalable Dynamic Credential Management in IoT,” IEEE Internet of Things Journal, vol. 12, no. 21, pp. 45663–45679, Nov. 2025, doi: 10.1109/JIOT.2025.3600371
  5. D. Díaz-Sánchez, F. Almenárez-Mendoza, C. Campo-Vázquez, y C. García-Rubio, “Zero-trust token authorization with trapdoor hashes for scalable distributed firewalls,” Future Generation Computer Systems, vol. 176, art. no. 108227, 2026, doi: 10.1016/j.future.2025.108227
  6. Alberto Bazán-Guillén, Carlos Beis-Penedo, Diego Cajaraville-Aboy, Pablo Barbecho-Bautista, Rebeca P. Díaz-Redondo, Luis J. de la Cruz Llopis, Ana Fernández-Vilas, Mónica Aguilar Igartua, Manuel Fernández-Veiga, «Realistic Urban Traffic Generator using Decentralized Federated Learning for the SUMO simulator», IEEE Open Journal of the Communications Society, ISSN: 2644-125X, 8th August 2025, (IF 2024 = 6.1; 23/120; Telecommunications; Q1), DOI: 10.1109/OJCOMS.2025.3597019
  7. Yaqoob Al-Zuhairi, Prashanth Kannan, Alberto Bazán Guillén, Luis J. de la Cruz Llopis, Mónica Aguilar Igartua, «Efficient Charging Station Selection for Minimizing Total Travel Time of Electric Vehicles», Future Internet, ISSN: 2644-125X, 18th August 2025, (IF 2024 = 3.6; 93/258; Computer Science, Information Systems; Q2), DOI: 10.3390/fi17080374
  8. Gerard Caravaca Ibáñez, Luis J. de la Cruz Llopis, Adrián Catalín Diaconeasa, Alberto Bazán Guillén, Mónica Aguilar Igartua, «MobilitApp: A Deep Learning-Based Tool for Transport Mode Detection to Support Sustainable Urban Mobility», IEEE Access, ISSN: 2169-3536, 13th April 2025, (IF 2024 = 3.6; 50/120; Telecommunications; Q2), DOI: 10.1109/ACCESS.2025.3561238
  9. Antoni Mínguez, Volkan Tozan, Monica El-Assaad, Silvia Solà-Muñoz, Youcef Azeli, Luis J. de la Cruz Llopis, Xavier Jiménez-Fàbrega, Carles Galup, Raimon Dalmau, Mónica Aguilar Igartua, «Design of a SUMO-based simulator for optimal location of emergency vehicles in the Emergency Medical Systems», Emergency Medicine International, ISSN: 2090-2840, 4823481, pp. 1-26, February 2025, (IF 2024 = 0.8; 39/56; Emergency Medicine; Q3), DOI: 10.1155/emmi/4823481
  10. Tobar Nicolau, A., Parra-Arnau, J.*, Forné, J., Torra, V. (2025): Uncoordinated syntactic privacy: a new metric for multiple, independent data publishing. In: IEEE Transactions on Information Forensics and Security, vol. 20, pp. 3362-3373, IF (2024): 8.0; 12/147, D1 (computer science, theory & methods). DOI: 10.1109/TIFS.2025.3551645
  11. Pablo Fernández-Piñeiro, M. Fernández-Veiga, R.P., Díaz-Redondo, A. Fernández-Vilas, M. González-Soto. “Towards efficient compression and communication for prototype-based decentralized learning”. Applied Soft computing, 2025. DOI: 10.1016/j.asoc.2025.113270. Also in arXiv: 2411.09267
  12. X. Martínez Luaña, Manuel Fernández-Veiga, Rebeca P. Díaz-Redondo. “Privacy-aware Berrut Approximated Coded Computing for Federated Learning”. J. Network and Computer Applications, 2025. DOI: 10.1016/j.jnca.2025.104280
  13. Diego Cajaraville-Aboy, Ana Fernández-Vilas, Rebeca Díaz-Redondo, Manuel Fernández-Veiga. “Byzantine-Robust Aggregation for Securing Decentralized Federated Learning”. IEEE Access, 2025. DOI: 10.1109/ACCESS.2025.3629864
  14. Martín González-Soto, Bruno Fernández-Castro, Rebeca P. Díaz-Redondo, M. Fernández-Veiga. “IC4Net: Decentralized Communication for Continual Multi-Agent Learning”. IEEE Access, 2025. DOI: 0.1109/ACCESS.2025.3631242
  15. C. Beis-Penedo, R. P. Díaz-Redondo, A. Fernández-Vilas, M. Fernández-Veiga. “A Blockchain Solution for Decentralized Training in Machine Learning for IoT”. Computer Communications, 2025. DOI: 10.1016/j.comcom.2025.108289
  16. Diego Cajaraville-Aboy, Marta Moure-Garrido, Carlos Beis-Penedo, Carlos Garcia-Rubio, Rebeca P. Díaz-Redondo, Celeste Campo, Ana Fernández-Vilas, and Manuel Fernández-Veiga. «CO-DEFEND: Continuous Decentralized Federated Learning for Secure DoH-Based Threat Detection». Computer Networks, 2025. arXiv: arXiv: 2504.01882 DOI: 10.1016/j.comnet.2025.111961
  17. C. Beis-Penedo, Rebeca P. Díaz-Redondo, A. Fernández-Vilas, M. Fernández-Veiga. «HLF-FSL: A Decentralized Federated Split Learning Solution for IoT on Hyperledger Fabric». Array, 2026. DOI: 10.1016/j.array.2026.100685
  18. Álvaro Rodríguez-Vázquez, Carlos Giraldo, M. Fernández-Veiga. «CNFP: Optimizing Cloud-Native Network Function Placement with Diffusion Models on the Cloud Continuum». Computer Networks, 2026. DOI: 10.1016/j.comnet.2026.112321
  19. Diego Cajaraville-Aboy, Ana Fernández-Vilas, Rebeca P. Díaz-Redondo, Manuel Fernández-Veiga, PhD; Pablo Picallo-López. «Decentralized Orchestration Architecture for Fluid Computing: A Secure Distributed AI Use Case». Computer Networks, 2026.

News

VIEW ALL

Contact

Escola de Enxeñaría de Telecomunicación

Rúa Maxwell, s/n, 36310 Vigo, Pontevedra

X-twitter Linkedin
Escola de enxeñería de telecomunicación
UVigo
  • iclab@uvigo.gal
  • +34 986813868
  • Rúa Maxwell, s/n, 36310 Vigo