Member in the Spotlight: Riccardo Lazzeretti

In this series, we aim to introduce society members and other experts of the signal processing field. This month, we are happy to introduce Riccardo Lazzeretti, Post Doc from Information Engineering and Mathematics Department of the University of Siena. His research activity is mainly focused on privacy-preserving applications based on secure two-party computation tools.

What is your application that requires signal processing technologies?

The first time I approached signal processing technologies was while working on my master thesis, where I had to compress encrypted images by using cryptographic and signal processing instruments. Afterwards I kept on researching in the field of Signal Processing in the Encrypted Domain (SPED). This research field links signal processing and cryptographic communities and aims to solve privacy preserving problems. The cryptographic community is working hardly on Secure Multi-Party Computation (SMPC) tools for processing data while encrypted. On the other side, despite man improvements, the complexity of the SMPC tools is so high that it is impossible to release efficient protocols. Only by performing an accurate signal processing analysis it is possible to decrease significantly the complexity and develop efficient protocols. In the specific I implemented privacy preserving protocols which permit to classify encrypted ElectroCardioGrams, find the biometric in a database matching with the query, etc.

In your opinion, are there any emerging applications that signal processing technology may play an important role?

Big Data analysis is one of the hot topics of the latest years. However important privacy risks arise. I’m sure that SPED and signal processing advances can significantly contribute to mitigate the privacy-related problem in cloud computing or distributed applications

In your opinion, what are the most exciting NEW techniques in signal processing field?

I am impressed by the great advancements in biometric processing. New biometric signals are addressed. Some of them are considering new biometric traits, such as the ECG or the echo produced by the ear. Others approach human behavior, such as the way a person move the phone while grabbing it or the movements made during text composing. Also related problems are studied. Really interesting is the research carried on spoofing and anti-spoofing problems. On the light of my expertize, I am also amazed by the recent novelties in SMPC. The Gentry’s Fully Homomorphic Encryption scheme opened the possibility of performing any computation on encrypted data without interaction. However this still is a chimera, given the high protocol complexity, despite all the recent improvements. On the other side it is cleverly used in the pre-computation phase of the new SPDZ tools, allowing to develop efficient protocols secure against malicious parties. The use of such an instrument together with signal processing tools can spread the diffusion of SPED not only in the research community, but also in real world applications.

In which way have you been connected first with IEEE SPS (university, conference, etc…)?

I became a student IEEE member during my PhD studies, when my paper was accepted at the first Workshop on Information Forensic and Security. I am also a member of the SPS for two years.

Brief Bio

Riccardo Lazzeretti (lazzeretti[at]diism.unisi[]it) graduated with a degree in computer science engineering from the University of Siena, Italy, in 2007, where he continued his studies as a Ph.D. student under the supervision of Prof. Mauro Barni in the Information Engineering Department. From November 2009 to May 2010, he was with Philips Lab in Eindhoven, The Netherlands. In 2012, he received a research grant and continued his research in the Information Engineering and Mathematics Department of the University of Siena. His research activity is mainly focused on privacy-preserving applications based on secure two-party computation tools.