|
Henrique João Lopes Domingos
Departamento de Informática Office: P3/3, DI-FCT-UNL |
[Home] [Affiliations & Bio] [Teaching] [Research] [Ref. Projects] [Students] [Extensions]
My research is focused on Security and Privacy in the context of Large Scale Scale Distributed Systems, Computer Networks and IoT-Edge-Cloud Continuum Systems. My research topics cover the study of Dependability, Security, Trustworthy, Privacy-Preservation and Trusted Execution solutions for Distributed Computing Systems, Internetworking Protocols, Distributed Data-Management, Confidential Computing and Trusted Computing Systems.
I am particularly interested in finding solutions for (i) new cryptographic constructions for usable/practical privacy-preserved data-management solutions for outsourced dependable computing systems (ii) support for privacy-preserved data-flows and computations integrating user-applications and cyber-physical devices with edge-cloud continuums,(iii) programming support for trustworthy portable functions to run in different locations of outsourced computing environments, and (iv) design models for confidential computing and isolated hardware-assisted trusted execution environments (TEEs).
Most of my current research interests are oriented to understand issues of privacy preservation and data confidentiality for dependable distributed computing system. The goal is to understand how to enable such properties as complementary dimensions addressed by design, together with security, intrusion tolerance, reliability and availability guarantees.
To an even greater extent than with many other security, reliability and availability issues, my scientific understanding of new challenges lags far behind the need for rigorous defensive or preventive strategies. For a more effective reasoning about privacy challenges in the targeted distributed computing environments, I believe that we must also find novel solutions to support randomness, trustworthiness and verifiable trusted functions with fully decentralized trust computing models.
Furthermore, it is necessary to find ways to not limit the adversarial uncertainty, to preserve inference control in observable data-access, operations or adversarial correlation analysis of outsourced computations. To address this, several threads related to security and privacy topics in my research must be understood in multidisciplinary views. Currently I have a particular attention on how to support privacy-preserved machine learning with confidential computing arguments and what paradigms are particularly interesting for privacy-aware programming models and methodologies.