Pith. sign in

REVIEW

Generative AI-driven Cross-layer Covert Communication: Fundamentals, Framework and Case Study

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2501.11068 v1 pith:OFT5I5IU submitted 2025-01-19 cs.CR

Generative AI-driven Cross-layer Covert Communication: Fundamentals, Framework and Case Study

classification cs.CR
keywords communicationcovertcross-layernetworksecurityapplicationartificialcase
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
Share X Bluesky LinkedIn Reddit HN
read the original abstract

Ensuring end-to-end cross-layer communication security in military networks by selecting covert schemes between nodes is a key solution for military communication security. With the development of communication technology, covert communication has expanded from the physical layer to the network and application layers, utilizing methods such as artificial noise, private networks, and semantic coding to transmit secret messages. However, as adversaries continuously eavesdrop on specific communication channels, the accumulation of sufficient data may reveal underlying patterns that influence concealment, and establishing a cross-layer covert communication mechanism emerges as an effective strategy to mitigate these regulatory challenges. In this article, we first survey the communication security solution based on covert communication, specifically targeting three typical scenarios: device-to-device, private network communication, and public network communication, and analyze their application scopes. Furthermore, we propose an end-to-end cross-layer covert communication scheme driven by Generative Artificial Intelligence (GenAI), highlighting challenges and their solutions. Additionally, a case study is conducted using diffusion reinforcement learning to sovle cloud edge internet of things cross-layer secure communication.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.