Skip to content
Preprint

TwinViT-DeepJSCC: Adversarially Robust Semantic Image Communication

Oct 2026 · 1 citation · 42 references
Computer Science Engineering

TL;DR

TwinViT-DeepJSCC, a preventive-corrective semantic image transceiver operating under a fixed channel-use budget, achieves maximum peak signal-to-noise ratio gains and ablation results confirm the complementary contributions of the proposed transmitter- and receiver-side mechanisms.

Abstract

Learning-based semantic communication is vulnerable to adversarial perturbations introduced before semantic encoding or over wireless channels. This paper proposes TwinViT-DeepJSCC, a preventive-corrective semantic image transceiver operating under a fixed channel-use budget. Two Vision Transformer (ViT)-based deep joint source-channel coding (DeepJSCC) branches learn complementary latent representations protected by sensitivity-aware masking. At the receiver, confidence-aware fusion, blind corruption-severity estimation, and signal-to-noise ratio (SNR)-severity-conditioned denoising diffusion implicit model (DDIM) purification mitigate residual corruption without requiring attack metadata. Experiments on the Canadian Institute for Advanced Research 100-class (CIFAR-100) dataset consider fast gradient sign method (FGSM), projected gradient descent (PGD), natural evolution strategies (NES), and Carlini-Wagner (CW) source-domain attacks, as well as random jamming and channel-aware adversarial waveforms over additive white Gaussian noise (AWGN) and block-flat Rayleigh fading. Under matched channel-use and attack budgets, TwinViT-DeepJSCC achieves maximum peak signal-to-noise ratio (PSNR) gains of approximately 9.5 dB under 20-step PGD and 10.8 dB under channel-aware waveform attacks over block-flat Rayleigh fading. Under PGD, it also improves Top-1 accuracy by up to approximately 38 percentage points over the undefended baseline and 13 percentage points over the strongest competing defense. Ablation results confirm the complementary contributions of the proposed transmitter- and receiver-side mechanisms.

View source

Similar papers

Preprint Sep 2026

Where to Defend? Layer-Wise Adversarial Training for Robust Transformer-Based Semantic Communications

Deep learning-based semantic communication (DeepSC), a Transformer-based encoder-decoder, achieves semantic fidelity over noisy channels but remains vulnerable to adversarial perturbations injected at multiple stages of the pipeline. We present a layer-wise robustness framework that compares fast gradient sign method (...

Maria Slim, Razane Tajeddine, Mariette Awad et al. · 0 citations
Preprint Sep 2026

Lightweight Generative Image Semantic Communication over Packet Erasure Channels

The proposed LGSemCom framework is a lightweight generative packet-level joint source-channel coding (JSCC) framework for efficient and robust image transmission over packet erasure channels that employs a fully convolutional codec based on efficient long-range attention blocks that capture global semantic dependencies...

Yu-Fei Bo, Hao-Shuo Zhang, Mei-Xia Tao et al. · 0 citations
Conference Sep 2026

SemCom-UAP: Universal Adversarial Perturbations Against Cross-Modal Retrieval in Semantic Communications

Semantic communication (SemCom) systems based on vision-language pre-training (VLP) models transmit compact embeddings for downstream tasks such as cross-modal retrieval, but also introduce new vulnerabilities to universal adversarial perturbations (UAPs). Existing VLP-targeted UAP methods typically assume noise-free t...

Zi-Yu Wang, Yi-Ming Du, Jian Li et al. · 0 citations
#machine learning Preprint Sep 2026

FedGenSC: Federated Generative Semantic Communication with Channel-Aware Adaptation

Federated generative semantic communication (FedGenSC) is proposed, which mitigates all three failure modes of naive GAN federation by employing a global generator with local-only discriminators, providing cross-client semantic information through a semantic prototype bank, and conditioning generation on the instantane...

Rita Abou Fares, Razan Al Kakoun, Maher Nouiehed et al. · 0 citations
#machine learning Preprint Sep 2026

Privacy-Preserving Deep Joint Source-Channel Coding with In-Loop Concept Erasure

LEAPSC (LEACE-in-the-loop privacy for semantic communication), whose core contribution is the integration of in-loop least-squares concept erasure (LEACE) within a variational information bottleneck (VIB) encoder, improves robustness across channel conditions and over the high-leakage tail of samples.

Rami Eid, Maria Slim, Mariette Awad et al. · 0 citations
Open access Aug 2026

GAN-Based Semantic Communication for Image Transmission in IoV

For cooperative perception in the internet of vehicles, this paper proposes a generative adversarial network-based semantic communication framework to address the efficiency and fidelity bottlenecks of traditional communication systems in visual data transmission under limited bandwidth and dynamic channel conditions....

Rui-Xing Ren, Shan Chen, Jun-Hui Zhao et al. · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.