Scholarship 24/12903-0 - Aprendizado federado, Internet das coisas - BV FAPESP
Advanced search
Start date
Betweenand

Federated device fingerprinting and identification for enhanced intrusion detection in evolving IoT systems

Grant number: 24/12903-0
Support Opportunities:Scholarships abroad - Research Internship - Doctorate
Start date until: March 01, 2025
End date until: August 31, 2025
Field of knowledge:Engineering - Electrical Engineering - Telecommunications
Principal Investigator:João Henrique Kleinschmidt
Grantee:Ogobuchi Daniel Okey
Supervisor: Sajjad Dadkhah
Host Institution: Centro de Engenharia, Modelagem e Ciências Sociais Aplicadas (CECS). Universidade Federal do ABC (UFABC). Ministério da Educação (Brasil). Santo André , SP, Brazil
Institution abroad: University of New Brunswick, Fredericton (UNB Fredericton), Canada  
Associated to the scholarship:23/07184-1 - Enhancing the privacy and security of IoT intrusion detection systems through federated learning, BP.DR

Abstract

This project proposal seeks to address the critical need for scalable and privacy-oriented intrusion detection in the rapidly expanding Internet of Things (IoT). Classical centralized methods have shown inefficiencies in view of the sheer volume and heterogeneity of IoT devices. To arrest these challenges, device fingerprints will be extracted using behavioural and statistical information, which are then applied to develop a federated model to identify devices based on the fingerprinted information. The process is then integrated into the intrusion detection systems (IDS), which helps to ensure that only secure devices contribute their information (data, power, weights) to the IDS. Through proactive isolation of potentiallycompromised devices identified in the network from engaging in data transmission, the risksof exposing all devices to threats are eliminated. With the introduction of federated learningarchitecture, an assurance is established on the privacy of device information while the fingerprints are extracted. The proposed system will involve two modules: device fingerprinting and identification. Ultimately, the identified features are employed to train the device identification model using suitable machine learning algorithms. As part of a larger PhD project focused on optimizing privacy-centric and communication-efficient IDS for IoT, this research will contribute to detecting and isolating compromised devices in federated learning environments. By ensuring only trustworthy devices participate in the learning process, the proposed system will enhance the robustness of the PhD project in terms of accuracy, reliability, and security of IoT-based IDS. Additionally, by ensuring that only identified and permitted devices contribute to the IDS building process, we address the challenge of communication bottleneck as the available network bandwidth will not be consumed by unsolicited devices.

News published in Agência FAPESP Newsletter about the scholarship:
More itemsLess items
Articles published in other media outlets ( ):
More itemsLess items
VEICULO: TITULO (DATA)
VEICULO: TITULO (DATA)

Please report errors in scientific publications list using this form.