| Author: | Li, Hao |
| Title: | Byzantine-robust time series data release under local differential privacy |
| Degree: | M.Sc. |
| Year: | 2023 |
| Department: | Department of Electrical and Electronic Engineering |
| Pages: | 1 volume (unpaged) : color illustrations |
| Language: | English |
| Abstract: | Local differential privacy is one of the concepts in differential privacy. Compared with traditional differential privacy, local differential privacy emphasizes the realization of data privacy and security through local users' own disturbance of data. To put it simply, the user perturbs the data locally and then sends the perturbed data to the server. The server then obtains effective information through data aggregation. Doing so protects the user's information if it is stolen by an attacker. Nowadays, many local differential privacy protocols have been proposed, and there has been a lot of research on local differential privacy. However, there is still a lot of work to be done on the security of local differential privacy protocols. In this work, our main goal is to discuss the security of time series data under LDP protocols. We focus on the KRR, OUE and OLH protocols, which are basic LDP protocols. We show the theory of these protocols, and we use these protocols to do the data analytics work, which are frequency estimation and heavy hitter identification. Then we use the data poisoning attacks to attack these protocols by inject fake users, and we show the attack effect of these attacks. We simulated the attack effect of data poisoning attack on estimate frequency and heavy hitter detection under LDP protocols. We also proposed two defense methods against these attacks and conducted experiments under different LDP protocols and under different parameters to prove the practicability of these two methods. Finally, we summarized our experimental data and discussed the next work direction. |
| Rights: | All rights reserved |
| Access: | restricted access |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| 8280.pdf | For All Users (off-campus access for PolyU Staff & Students only) | 2.49 MB | Adobe PDF | View/Open |
Copyright Undertaking
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
- I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.
Please use this identifier to cite or link to this item:
https://theses.lib.polyu.edu.hk/handle/200/13875