Summary
"The world has become one global village, strongly connected by the internet. Users, applications, and companies produce huge amounts of data. This data is mostly very private (health, financial, commercial data, etc.), requiring organizations to encrypt the data when storing or transmitting it. In Europe, fortunately, our GDPR regulations are very strict about this. On the other hand, many legitimate applications want to inspect or match the data with some database. Examples are matching encrypted names with a red list, providing privacy in financial transactions, or matching encrypted DNA snippets in a database.
A solution to these conflicting requirements is offered by a new field in cryptography: computing on encrypted data (COED), and more specifically, fully homomorphic encryption (FHE). With FHE, data sits encrypted in the cloud. The data holder, i.e. the remote server, has no knowledge and is unable to decrypt the data. The magic of FHE is that operations can still be performed on the data, even though it remains encrypted. Unfortunately, it is currently too slow and too cumbersome to be used in practice. Within our ERC grant BELFORT, which focuses on ""Hardware Acceleration for Computing on Encrypted Data"", a novel hardware processor demonstrates orders of magnitude speed-up for essential operations in FHE.
This POC aims to move from our academic proof of concept into a full-stack project integrated into open-source and commercial FHE libraries. We then plan to run a demo on an application of database look-up operations (‘keyword matching’), as they appear in many FHE applications.
The ultimate goal of this project is to make operations on encrypted data efficient enough to perform them on the fly in a cloud setting. When a user formulates privacy-preserving encrypted queries to a server, it should be imperceptible that the operations are occurring in the encrypted domain. Being able to answer a query on the fly without delay will make this goal happen."
A solution to these conflicting requirements is offered by a new field in cryptography: computing on encrypted data (COED), and more specifically, fully homomorphic encryption (FHE). With FHE, data sits encrypted in the cloud. The data holder, i.e. the remote server, has no knowledge and is unable to decrypt the data. The magic of FHE is that operations can still be performed on the data, even though it remains encrypted. Unfortunately, it is currently too slow and too cumbersome to be used in practice. Within our ERC grant BELFORT, which focuses on ""Hardware Acceleration for Computing on Encrypted Data"", a novel hardware processor demonstrates orders of magnitude speed-up for essential operations in FHE.
This POC aims to move from our academic proof of concept into a full-stack project integrated into open-source and commercial FHE libraries. We then plan to run a demo on an application of database look-up operations (‘keyword matching’), as they appear in many FHE applications.
The ultimate goal of this project is to make operations on encrypted data efficient enough to perform them on the fly in a cloud setting. When a user formulates privacy-preserving encrypted queries to a server, it should be imperceptible that the operations are occurring in the encrypted domain. Being able to answer a query on the fly without delay will make this goal happen."
Unfold all
/
Fold all
More information & hyperlinks
| Web resources: | https://cordis.europa.eu/project/id/101158183 |
| Start date: | 01-03-2024 |
| End date: | 31-08-2025 |
| Total budget - Public funding: | - 150 000,00 Euro |
Cordis data
Original description
"The world has become one global village, strongly connected by the internet. Users, applications, and companies produce huge amounts of data. This data is mostly very private (health, financial, commercial data, etc.), requiring organizations to encrypt the data when storing or transmitting it. In Europe, fortunately, our GDPR regulations are very strict about this. On the other hand, many legitimate applications want to inspect or match the data with some database. Examples are matching encrypted names with a red list, providing privacy in financial transactions, or matching encrypted DNA snippets in a database.A solution to these conflicting requirements is offered by a new field in cryptography: computing on encrypted data (COED), and more specifically, fully homomorphic encryption (FHE). With FHE, data sits encrypted in the cloud. The data holder, i.e. the remote server, has no knowledge and is unable to decrypt the data. The magic of FHE is that operations can still be performed on the data, even though it remains encrypted. Unfortunately, it is currently too slow and too cumbersome to be used in practice. Within our ERC grant BELFORT, which focuses on ""Hardware Acceleration for Computing on Encrypted Data"", a novel hardware processor demonstrates orders of magnitude speed-up for essential operations in FHE.
This POC aims to move from our academic proof of concept into a full-stack project integrated into open-source and commercial FHE libraries. We then plan to run a demo on an application of database look-up operations (‘keyword matching’), as they appear in many FHE applications.
The ultimate goal of this project is to make operations on encrypted data efficient enough to perform them on the fly in a cloud setting. When a user formulates privacy-preserving encrypted queries to a server, it should be imperceptible that the operations are occurring in the encrypted domain. Being able to answer a query on the fly without delay will make this goal happen."
Status
SIGNEDCall topic
ERC-2023-POCUpdate Date
12-03-2024
Images
No images available.
Geographical location(s)
