Domain-Adaptive Deep Joint Source-Channel Coding for Image Classification

2026-07-30

AI Summary

This work addresses the performance degradation of deep joint source-channel coding (Deep JSCC) in image classification under distribution shifts between training and deployment domains. To mitigate this issue, the authors propose a domain-adaptive Deep JSCC framework that integrates pseudo-label-guided class-level adversarial alignment with confidence-filtered supervised contrastive learning, enabling end-to-end optimization over both AWGN and Rayleigh fading channels without requiring additional inference networks. A novel classification–capacity–invariance (CCI) function is introduced to characterize the interplay between channel capacity, cross-domain invariance, and classification accuracy. Evaluated on the SVHN→MNIST transfer task at 10 dB CSNR, the method achieves a target-domain accuracy of 98.15%, substantially outperforming existing baselines.

Abstract

Deep joint source--channel coding (Deep JSCC) enables visual semantic transmission by mapping inputs directly to channel symbols and task outputs, but its performance can deteriorate under distribution shifts between training and deployment domains. We study single-source domain adaptation for task-oriented Deep JSCC and formulate a classification-capacity-invariance (CCI) function to characterize how the available channel capacity and class-conditional cross-domain invariance affect target domain classification accuracy. A scalar linear analysis of source-domain-optimal solutions and a controlled shallow nonlinear validation show that target domain classification accuracy can vary non-monotonically with the invariance constraint and with available capacity along separate control paths obtained by varying the transmitted dimension or CSNR. We then propose a domain-adaptive Deep JSCC framework that combines pseudo-label-based class-level adversarial alignment with supervised contrastive learning on confidence-filtered target samples. Experiments on digit and PACS datasets over AWGN and Rayleigh fading channels demonstrate improved target domain generalization without introducing additional inference-time networks. On SVHN $\rightarrow$ MNIST, the proposed method achieves 98.15\% target-domain accuracy at a CSNR of 10 dB.

Problem

• domain adaptation
• deep joint source-channel coding
• distribution shift
• image classification
• target domain generalization

Innovation

• domain adaptation
• deep joint source-channel coding
• classification-capacity-invariance
• adversarial alignment
• supervised contrastive learning