contINuous deCentralized lEarNing of ioT devIces behaVioural profilEs
In an increasingly hyperconnected society, cybersecurity concerns take on a broader dimension, which can impact citizens’ safety. This has been widely recognized in the EU through specific legal instruments, such as the Cybersecur...
ver más
28/02/2025
UM
181K€
Presupuesto del proyecto: 181K€
Líder del proyecto
UNIVERSIDAD DE MURCIA
No se ha especificado una descripción o un objeto social para esta compañía.
Total investigadores911
Fecha límite participación
Sin fecha límite de participación.
¿Tienes un proyecto y buscas un partner? Gracias a nuestro motor inteligente podemos recomendarte los mejores socios y ponerte en contacto con ellos. Te lo explicamos en este video
Información proyecto INCENTIVE
Duración del proyecto: 31 meses
Fecha Inicio: 2022-07-14
Fecha Fin: 2025-02-28
Líder del proyecto
UNIVERSIDAD DE MURCIA
No se ha especificado una descripción o un objeto social para esta compañía.
Total investigadores911
Presupuesto del proyecto
181K€
Fecha límite de participación
Sin fecha límite de participación.
Descripción del proyecto
In an increasingly hyperconnected society, cybersecurity concerns take on a broader dimension, which can impact citizens’ safety. This has been widely recognized in the EU through specific legal instruments, such as the Cybersecurity Act and the recent Cybersecurity Strategy for the Digital Decade. The development of the IoT technologies have fostered the deployment of data-driven services for everyday scenarios, such as transport, health and energy, but have also increased the probability and impact of new cybersecurity threats. Recent and well-known attacks have demonstrated the need to develop AI-based techniques to identify such attacks in IoT scenarios. In this context, the objective of this project is to build a decentralized framework to learn the IoT devices’ intended behavior throughout their lifecycle, in order to distinguish abnormal behavior that may reflect a security attack or threat. The proposal will build an edge-based and federated learning architecture to enable IoT devices to participate in the learning process by using lightweight security protocols. This architecture will include the use of blockchain through a novel approach in which different ledgers will be interconnected to improve the performance and practicability of existing approaches. Furthermore, the proposal will be the first effort to quantify the impact of human-machine interactions on the devices’ intended behavior. The project will be validated quantitatively and qualitatively to identify potential tradeoffs for its deployment in different IoT-enabled scenarios. It will leverage the candidate’s extensive knowledge and experience in IoT cybersecurity, which will be strengthened by the host institution to improve his technical and transferable skills in a multidisciplinary environment. The proposed project represents a unique contribution to reinforcing the culture of cybersecurity in the EU that will boost the candidate's professional development during and after the fellowship.