Garbled Circuit Generation for Private Set Intersection
Garbled Circuit Generation for Private Set Intersection
| Supervisor(s): | Mark Gall |
| Status: | finished |
| Topic: | Others |
| Author: | Erik Pohle |
| Submission: | 2018-08-15 |
| Type of Thesis: | Bachelorthesis |
Thesis topic in co-operation with the Fraunhofer Institute for Applied and Integrated Security AISEC, Garching
|
|
DescriptionMotivation and Topic Data analysis is widely used to gain useful information from vast data sets. Usage of this information varies widely and can include recommendation systems for streaminng services or navigation systems with realtime traffic information. When the data is owned by one party and the analysis is done by another party both parties need to cooperate to gain these insights. But this cooperation carries risks for both parties. The data analyst might copy large portions or even the entire data set, gaining more information than intended. Similarly, the data owner might learn details about the analytic algorithms, which have been developed by the data analyst and are his assets. Secure Computation offers a solution to this scenario, such that the two parties can compute basically any function without revealing either the input nor the details of the computation with each other. One of the most important protocols for this task are Yao’s Garbled Circuits. In this protocol one party creates the garbled circuit and the other party evaluates the circuit. The protocol requires as input a boolean circuit representing the function to be computed. The goal of this thesis is to first research the current state of the art in generating boolean circuits, especially in the context of Secure Computation with Yao’s Garbled Circuits. In a second step, one approach for generating boolean circuits ready for Secure Computation will be implemented further limited to circuits on set operations, such as intersection and complement.
• Programming skills (Java, Python) • Interest in Secure Computation, Cryptography • Ability to work self-directed and systematically
|
|