Topology-Preserving Deep Joint Source-Channel Coding for Semantic Communication

2026-03-17

AI Summary

Existing deep joint source-channel coding (DeepJSCC) approaches prioritize pixel-level fidelity while neglecting global topological structures—such as connectivity—that are critical for structural semantics in applications like autonomous driving. To address this limitation, this work proposes TopoJSCC, a novel framework that integrates topological data analysis into semantic communication. Specifically, it introduces a persistent homology regularizer within an end-to-end DeepJSCC architecture, explicitly enforcing topological consistency by minimizing the Wasserstein distance between the cubical persistence diagrams of the original and reconstructed images, as well as between the Vietoris–Rips persistence diagrams of latent features before and after channel transmission. Without side information, TopoJSCC effectively preserves global image structure and achieves significant improvements in both topological fidelity and PSNR under low signal-to-noise ratio and bandwidth-constrained conditions.

Technology Category

• Computer Vision: Learning & Optimization for CV
• Machine Learning: Deep Generative Models & Autoencoders
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics
• Systems and Infrastructure for Web, Mobile and WoT: Experiences and lessons learnt from Web-based algorithms and system deployments

Abstract

Many wireless vision applications, such as autonomous driving, require preservation of global structural information rather than only per-pixel fidelity. However, existing Deep joint source-channel coding (DeepJSCC) schemes mainly optimize pixel-wise losses and provide no explicit protection of connectivity or topology. This letter proposes TopoJSCC, a topology-aware DeepJSCC framework that integrates persistent-homology regularizers to end-to-end training. Specifically, we enforce topological consistency by penalizing Wasserstein distances between cubical persistence diagrams of original and reconstructed images, and between Vietoris--Rips persistence of latent features before and after the channel to promote a robust latent manifold. TopoJSCC is based on end-to-end learning and requires no side information. Experiments show improved topology preservation and peak signal-to-noise ratio (PSNR) in low signal-to-noise ratio (SNR) and bandwidth-ratio regimes.

Problem

• topology preservation
• deep joint source-channel coding
• semantic communication
• persistent homology
• structural information

Innovation

• topology preservation
• persistent homology
• DeepJSCC
• semantic communication
• Wasserstein distance