Abstract
The low-latency, high bandwidth capabilities promised by 5G, together with the diffusion of applications that require high computing power and, again, low latency (such as videogames), are probably the main reasons—though not the only one—that have led to the introduction of a new network architecture: Fog Computing, that consists in moving the computation services geographically close to where computing is needed. This architectural shift moves security and privacy issues from the Cloud to the different layers of the Fog architecture. In this scenario, IDSs are still necessary, but they need to be contextualized in the new architecture. Indeed, while on the one hand Fog computing provides intrinsic benefits (e.g., low latency), on the other hand, it introduces new design challenges. In this paper, we provide the following contributions: we analyze the possible IDS solutions that can be adopted within the different Fog computing tiers, together with their related deployment and design challenges; and, we propose some promising future directions, by taking into account the challenges left uncovered by the considered solutions.
Original language | English (US) |
---|---|
Title of host publication | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
Publisher | Springer [email protected] |
Pages | 59-75 |
Number of pages | 17 |
ISBN (Print) | 9783030233730 |
DOIs | |
State | Published - Jan 1 2019 |
Externally published | Yes |