Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Secure Semantic Communication for Image Transmission in the Presence of Eavesdroppers

Shunpu Tang, Chen Liu, Qianqian Yang, Shibo He, Dusit Niyato

TL;DR

An invertible neural network (INN)-based signal steganography approach that embeds channel input signals of a private image into those of a host image before transmission ensures that the original private image can be reconstructed from the received signals at the legitimate receiver, while the eavesdropper can only decode the information of the host image.

Abstract

Semantic communication (SemCom) has emerged as a key technology for the forthcoming sixth-generation (6G) network, attributed to its enhanced communication efficiency and robustness against channel noise. However, the open nature of wireless channels renders them vulnerable to eavesdropping, posing a serious threat to privacy. To address this issue, we propose a novel secure semantic communication (SemCom) approach for image transmission, which integrates steganography technology to conceal private information within non-private images (host images). Specifically, we propose an invertible neural network (INN)-based signal steganography approach, which embeds channel input signals of a private image into those of a host image before transmission. This ensures that the original private image can be reconstructed from the received signals at the legitimate receiver, while the eavesdropper can only decode the information of the host image. Simulation results demonstrate that the proposed approach maintains comparable reconstruction quality of both host and private images at the legitimate receiver, compared to scenarios without any secure mechanisms. Experiments also show that the eavesdropper is only able to reconstruct host images, showcasing the enhanced security provided by our approach.

Secure Semantic Communication for Image Transmission in the Presence of Eavesdroppers

TL;DR

An invertible neural network (INN)-based signal steganography approach that embeds channel input signals of a private image into those of a host image before transmission ensures that the original private image can be reconstructed from the received signals at the legitimate receiver, while the eavesdropper can only decode the information of the host image.

Abstract

Semantic communication (SemCom) has emerged as a key technology for the forthcoming sixth-generation (6G) network, attributed to its enhanced communication efficiency and robustness against channel noise. However, the open nature of wireless channels renders them vulnerable to eavesdropping, posing a serious threat to privacy. To address this issue, we propose a novel secure semantic communication (SemCom) approach for image transmission, which integrates steganography technology to conceal private information within non-private images (host images). Specifically, we propose an invertible neural network (INN)-based signal steganography approach, which embeds channel input signals of a private image into those of a host image before transmission. This ensures that the original private image can be reconstructed from the received signals at the legitimate receiver, while the eavesdropper can only decode the information of the host image. Simulation results demonstrate that the proposed approach maintains comparable reconstruction quality of both host and private images at the legitimate receiver, compared to scenarios without any secure mechanisms. Experiments also show that the eavesdropper is only able to reconstruct host images, showcasing the enhanced security provided by our approach.
Paper Structure (16 sections, 16 equations, 5 figures, 1 table)

This paper contains 16 sections, 16 equations, 5 figures, 1 table.

Table of Contents

  1. Introduction
  2. Problem Statement
  3. SemCom for Image Transmission
  4. Eavesdropping in SemCom
  5. Naive Decoding
  6. Model Inversion Attack (MIA)
  7. Proposed approach
  8. Secure SemCom with Signal Steganography
  9. INN-based Signal Steganography Module
  10. Training procedure and Loss function
  11. Simulation
  12. Implementation Details
  13. Reconstruction performance at Bob
  14. Reconstruction performance at Eve
  15. Visual Comparsion
  16. ...and 1 more sections

Figures (5)

  • Figure 1: Overview of the proposed secure SemCom with signal steganography.
  • Figure 2: Architecture of the INN-based signal steganography module, which consists of several invertible blocks.
  • Figure 3: PSNR and MS-SSIM of reconstructed images at Bob by DeepJSCC and the proposed secure SemCom approach, where the main link SNR varies from 0 to 20 dB.
  • Figure 4: PSNR of reconstructed images at Eve by DeepJSCC and the proposed secure SemCom approach with naive decoding and MIA eavesdropping, respectively, where SNR of the eavesdropping link varies from 0 to 20 dB.
  • Figure 5: Visual comparison of the host image and private image reconstruction at Bob and Eve, where the SNR of the main link and the eavesdropping link is set to 5 dB.