You are here: Resources > FIDIS Deliverables > Privacy and legal-social content > D13.1 Addendum: Identity and impact of privacy enhancing technologies > 
Terminology  Title:
LIMITS OF PRIVACY ENHANCING TECHNOLOGIES
 

 

Limits of Privacy Enhancing Technologies

The deliverable D13.1 lacks proper conclusions and they were omitted on purpose. It is an introductory document and the goal is to give an overview of the technologies and leave a lot of space for readers to think about particular technologies. 

We have split privacy into two distinct parts – application and communication privacy. All the technologies used to provide privacy work very well, when used on random data and random communication. Unfortunately, humans are predictable and their behaviour features patterns that can be used to defeat or mitigate privacy properties offered by various technologies. 

One can liken it to using strong cryptography with the same plain text all the time. It may be hard to decrypt the encrypted messages, but you soon realise that it is the same message being encrypted all over again. 

There has been a lot of work done on quantifying privacy properties for communication anonymity systems, as described in Chapter 3, but we still miss verification of the results on large datasets of real data, real traffic. We believe that current estimates are reasonably good and there will be no changes in the order of magnitude, but the equations might get simplified or constants changed. 

There is much less known about privacy in databases. There has been some research done in the area of medical databases, but we generally know much less about real properties of privacy enhancing technologies applied on databases or other types of stored data. One of the reasons is vagueness of threat model definitions and their relevancy to real-world systems. 

We will present more detailed results about the technological limits of privacy enhancing technologies in the subsequent deliverables of our workpackage, including results obtained from analysis of large real-world datasets. 

 

Terminology  wp13_1Add_final_01.sxw  
11 / 11