TOFS: Feature Selection with Swap Supervision for Semantic Communication over Fading Channels
Organizations: Graduate School of System Informatics, Kobe University, Kobe, Japan · Information Technology Center, Nara Women’s University, Nara, Japan
Abstract
Task-oriented semantic communication transmits task-relevant features to support receiver inference under limited communication resources and adverse channel conditions. For image classification, the usefulness of a feature depends on the other transmitted features and the distortion introduced by the wireless channel. Effective feature prioritization should therefore account for classification performance after transmission of the selected features. This paper proposes a Task-Oriented Feature Selection (TOFS) framework and method, which evaluates feature replacements using receiver classification loss. Loss differences confirmed under independent transmission samples provide pairwise ranking supervision for a selector network. The trained selector predicts priorities in a single forward pass. CIFAR-10 simulations demonstrate improved classification accuracy over an importance-based feature selection baseline at the same transmitted feature count and average transmit power. The proposed TOFS framework shows 84% accuracy performance under Rayleigh fading and Gaussian noise, compared with the 68% performance of a SOTA method.
Figures & tables
| Item | Setting |
|---|---|
| Link and receiver | , Rayleigh + noise, dB; ZF |
| Continuation | 30 epochs, Adam, batch 256; learning rate with cosine decay; weight decay |
| Swap feedback | pairs; independent transmission instances; , |
| Test randomness | Channel seeds 2026–2028; permutation seed 7202026 |
| SNR sweep | dB |
| Method | Clean (%) | Rayleigh (%) |
|---|---|---|
| Full | 96.94 | |
| Random | 73.09 | |
| PLV | 96.82 | |
| TOFS | 96.81 |
| Out | In | Acc. (%) | ||
|---|---|---|---|---|
| B4 | T4+R4 | 32 | 84.00 | 0.90 |
| B8 | T8+R8 | 128 | 84.39 | 1.00 |
| B16 | T16+R16 | 512 | 84.20 | 1.43 |
| R8 | T8+R8 | 128 | 83.71 | 0.99 |
| T8 | T8+R8 | 128 | 65.38 | 0.99 |
| B8 | R16 | 128 | 84.25 | 0.99 |
Feature Selection
Feature selection identifies compact, informative variables or input dimensions to improve prediction, efficiency, interpretability, and statistical inference.
See all 213 papers