Research map / Label-efficient and robust learning / Test-time adaptation
RFF-TTA: Physical Information-Aware Prototype for Temporally Varying RF Fingerprinting Online Test-Time-Adaptation
Taotao Li, Yiyang Li, Zhenyu Wen, Jiahao Lin, Jinhao Wan, Jie Su, Cong Wang, Zhen Hong
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use RFF-TTA instead of CLPS | RFF recognition (ResNet-18) · RFF E1 -> D4 | accuracy: 0.6125 -> 0.8091 | Improved |
| use RFF-TTA instead of CLPS | RFF recognition (ResNet-18) · RFF E1 -> D3 | accuracy: 0.7624 -> 0.9521 | Improved |
| use RFF-TTA instead of CLPS on same-day data | RFF recognition (ResNet-18) · RFF E1 -> D1 | accuracy: 0.9750 -> 0.9630 | Hurt |
| add stable RFF features (M1) (ablation) | RFF-TTA · RFF Day6 test | accuracy: 0.8101 -> 0.9153 | Improved |
| add CFO augmentation (M2) (ablation) | RFF-TTA · RFF Day7 test | accuracy: 0.8762 -> 0.8952 | Improved |
| add CFO augmentation (M2) (ablation) | RFF-TTA · RFF Day6 test | accuracy: 0.9153 -> 0.9195 | No clear effect |
| add CPO augmentation (M3) (ablation) | RFF-TTA · RFF Day6 test | accuracy: 0.9195 -> 0.9512 | Improved |
| add transient features (M4) (ablation) | RFF-TTA · RFF Day6 test | accuracy: 0.9512 -> 0.9674 | Improved |
| add transient features (M4) (ablation) | RFF-TTA · RFF Day9 test | accuracy: 0.8944 -> 0.9355 | Improved |
| use CFO/CPO physical augmentation instead of Gaussian noise (ablation) | RFF-TTA · RFF Day2 test | accuracy: 0.8465 -> 0.9118 | Improved |
| use CFO/CPO physical augmentation instead of time warp (ablation) | RFF-TTA · RFF Day3 test | accuracy: 0.8330 -> 0.9475 | Improved |
| apply standard CFO compensation | RFF pipeline · RFF multi-day test | accuracy | Hurt |
Closest papers by meaning
- Open-world Radio Frequency Fingerprint Identification via Augmented Semi-supervised Learning (AAAI 2025 · similarity 0.87)
- TinyTTA: Efficient Test-time Adaptation via Early-exit Ensembles on Edge Devices (NeurIPS 2024 · similarity 0.78)
- Blind-Touch: Homomorphic Encryption-Based Distributed Neural Network Inference for Privacy-Preserving Fingerprint Authentication (AAAI 2024 · similarity 0.78)
- Scalable Fingerprinting of Large Language Models (NeurIPS 2025 · similarity 0.78)
- Multitarget Device-Free Localization via Cross-Domain Wi-Fi RSS Training Data and Attentional Prior Fusion (AAAI 2024 · similarity 0.78)
- Intersecting-Boundary-Sensitive Fingerprinting for Tampering Detection of DNN Models (ICML 2024 · similarity 0.77)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
