Developing AI attacks/defenses
Mobile Edge Computing (MEC) is a promising technology that brings computation and storage resources closer to end-users, enabling low-latency, high-bandwidth, and energy-efficient processing of data at the edge of the cellular network. This technology has the potential to support a wide range of app...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project |
| Language: | English |
| Published: |
Nanyang Technological University
2023
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/165870 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Nanyang Technological University |
| Language: | English |
| Summary: | Mobile Edge Computing (MEC) is a promising technology that brings computation and storage resources closer to end-users, enabling low-latency, high-bandwidth, and energy-efficient processing of data at the edge of the cellular network. This technology has the potential to support a wide range of applications, such as smart cities, industrial automation, healthcare, and autonomous vehicles, by providing real-time processing of data. MEC face several challenges, including resource management, privacy, security, and scalability, which has been addressed to some extent by researchers. However, security for Artificial Intelligence (AI) models used in MEC is a field that is lacking proper research. This can potentially lead to a disastrous impact on end-users that may be targeted by adversaries. In this paper, we are summarizing the latest advancements of MEC security and the potential impact of state-of-the-art attacks on MEC models. We discuss the usage of MEC processes and their potential attacks with suggested defences. Finally, we propose a case study where we show an example of how an adversary can negatively affect a MEC model. |
|---|