Theoretically-Grounded Privacy-Preserving Deep Semantic Coding

doi:10.1109/LSP.2025.3631425

Theoretically-Grounded Privacy-Preserving Deep Semantic Coding

2025-01-01

0

PRE

AI

DOI：10.1109/LSP.2025.3631425

Original

Original request for help

Save

Abstract

En 中文

As an emerging data transmission paradigm, semantic communication can significantly improve transmission efficiency by focusing on extracting and preserving task-critical semantic information. However, privacy preservation of sensitive information in semantic communication has not been well studied. This letter focuses on the semantic coding and studies the fundamental trade-off among data compression, source distortion, semantic distortion, and privacy preservation from an information-theoretic perspective. Based on the theoretic results, a novel deep learning method is proposed for data-driven privacy-preserving semantic coding design. The theory and method are validated through experiments on the CelebA dataset.

Keywords:

Information-theoretic privacy

lossy compression

semantic communication

variational method

Theoretically-Grounded Privacy-Preserving Deep Semantic Coding

Abstract

Journal

Organization