Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Unveiling Covert Semantics: Joint Source-Channel Coding Under a Covertness Constraint

Abdelaziz Bounhar, Mireille Sarkiss, Michèle Wigger

TL;DR

The fundamental limit of Semantic Communications (joint source-channel coding) is established when the transmission needs to be kept covert from an external warden and it is shown that source and channel coding separation holds for this setup.

Abstract

The fundamental limit of Semantic Communications (joint source-channel coding) is established when the transmission needs to be kept covert from an external warden. We derive information-theoretic achievability and matching converse results and we show that source and channel coding separation holds for this setup. Furthermore, we show through an experimental setup that one can train a deep neural network to achieve covert semantic communication for the classification task. Our numerical experiments confirm our theoretical findings, which indicate that for reliable joint source-channel coding the number of transmitted source symbols can only scale as the square-root of the number of channel uses.

Unveiling Covert Semantics: Joint Source-Channel Coding Under a Covertness Constraint

TL;DR

The fundamental limit of Semantic Communications (joint source-channel coding) is established when the transmission needs to be kept covert from an external warden and it is shown that source and channel coding separation holds for this setup.

Abstract

The fundamental limit of Semantic Communications (joint source-channel coding) is established when the transmission needs to be kept covert from an external warden. We derive information-theoretic achievability and matching converse results and we show that source and channel coding separation holds for this setup. Furthermore, we show through an experimental setup that one can train a deep neural network to achieve covert semantic communication for the classification task. Our numerical experiments confirm our theoretical findings, which indicate that for reliable joint source-channel coding the number of transmitted source symbols can only scale as the square-root of the number of channel uses.
Paper Structure (10 sections, 1 theorem, 12 equations, 4 figures, 3 tables)

This paper contains 10 sections, 1 theorem, 12 equations, 4 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Information-Theoretic Approach
  3. Notation
  4. Problem statement
  5. Information-Theoretic Results
  6. Covert Semantic Extraction: Neural Network Experiment
  7. DNN Architectures and Training
  8. Numerical Results
  9. Summary and Discussion
  10. Information-Theoretic Converse Proof

Key Result

Theorem 1

For any given function $f(\cdot)$ and vanishing sequence $\{\delta_n \}_{n\geq 1}$ the following holds. Notice that the parameter $\gamma$ plays the same role as the bandwidth mismatch factor in traditional JSCC.

Figures (4)

  • Figure 1: Covert semantic communication system.
  • Figure 2: Separate source and channel coding architecture.
  • Figure 3: Neural Network architecture for distributed classification under a covertness constraint.
  • Figure 4: Accuracy in (%) as a function of the blocklength $n$ for the three models at SNR=1dB. The solid red curve denotes the Non-covert model, the dashed blue curve the Square-root covert model and the dash-dotted black curve the Linear covert model.

Theorems & Definitions (2)

  • Definition 1
  • Theorem 1