Oblivious NFA evaluation with an untrusted third party
Oblivious substring matching is a well-studied problem, with plentiful applications in scenarios with privacy-preservation requirements, e.g., genome matching, malicious code detection, etc. A natural generalization of oblivious substring matching is oblivious finite automata evaluation, which matc...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project |
| Language: | English |
| Published: |
Nanyang Technological University
2022
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/162890 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Nanyang Technological University |
| Language: | English |
| Summary: | Oblivious substring matching is a well-studied problem, with plentiful applications in scenarios with privacy-preservation requirements, e.g., genome matching, malicious code detection, etc. A natural generalization of oblivious substring matching is oblivious finite automata evaluation, which matches a wider collection of patterns than substrings. In this paper, we propose an efficient protocol for the oblivious evaluation of Non-deterministic finite Automata (NFA) with the help of an untrusted third party and extend the protocol to Non-deterministic Finite state transducers. We prove the security of the protocol against one corrupted party. We further experiment with such protocol on a pattern-matching task with its complexity. |
|---|