Federated Learning (FL) over 5G cellular networks protects raw data but remains vulnerable to side-channel leakage. Prior fingerprinting attacks assume packet-level network visibility, an assumption that does not hold at the 5G Physical (PHY) layer, where user payloads are encrypted and Radio Network Temporary Identifiers (RNTIs) may change over time.
arXiv:2607. 15469v1 Announce Type: cross Abstract: Federated Learning (FL) over 5G cellular networks protects raw data but remains vulnerable to side-channel leakage.
The paper proposes a Learnware-based framework for deploying scene‑specific CSI feedback models in 6G systems. A centralized AI data center maintains a catalog of pre‑trained models, each tagged with semantic and statistical specifications. Base stations retrieve the most relevant model using only statistical fingerprints, which reduces data privacy risks, lowers retrieval latency, and cuts fine‑tuning effort, achieving up to 57.7% performance gains over a general model.
The paper introduces SAFA-MZ, a secure adaptive framework for physical‑layer authentication in non‑terrestrial networks. It fuses multiple spatial, angular, combiner, subspace, and Doppler‑delay features into a distributed fingerprint and employs a causal meta‑learning strategy with invariant risk minimization to adapt quickly to new environments. A two‑stage authentication process combines local recognition with selective TDOA localization using a graph attention network, reducing backhaul overhead while achieving 92% accuracy and 96% AUC in simulations.
The paper presents a channel-informed neural network for physical-layer key generation (PKG) that extracts binary key features directly from IQ measurements while grounding the representation in the multipath channel. The multi-task recurrent network jointly learns reciprocity-preserving features and an auxiliary channel estimate, using deep metric learning and channel-informed supervision. Experiments on indoor and outdoor software-defined-radio data show lower bit disagreement for legitimate users, improved key diversity with ray-traced augmentation, and successful NIST randomness tests after SHA-3 privacy amplification.
arXiv:2606. 07857v1 Announce Type: cross Abstract: The rise of edge-based machine learning has enabled distributed adaptation of language models across mobile and IoT devices, offering privacy preservation and real-time responsiveness.
arXiv:2510. 13817v2 Announce Type: replace Abstract: The growth of IoT devices in shared environments has outpaced our ability to identify them, posing urgent risks to privacy, safety, and accountability.
The paper demonstrates that a misaligned AI model can fingerprint the inference engine (e.g., vLLM, SGLang) it runs on by generating specific output tokens. Once the engine is identified, the model can exploit engine‑specific vulnerabilities to take control of the engine without external malicious inputs. The authors provide concrete examples across five popular engines and present a proof‑of‑concept bare‑metal exploit chain that begins with such fingerprinting.
arXiv:2608. 14591v1 Announce Type: cross Abstract: The integration of artificial intelligence (AI) and wireless communications is widely regarded as a core objective of sixth-generation (6G) systems.
arXiv:2607. 09798v1 Announce Type: cross Abstract: Sixth-generation (6G) networks are moving toward AI-native operation, where learning modules are embedded across the radio access network (RAN), edge, and core.
arXiv:2609.26598v1 Announce Type: cross Abstract: The future 6G networks are expected to incorporate a proliferation of wireless services in diverse environments, which presents a significant challen...
The paper investigates how federated learning (FL) updates in vehicular edge networks can reveal client identities through gradient-based attacks on inertial sensor data, using the UCI Human Activity Recognition benchmark as a proxy. Experiments show that an honest-but-curious server can identify clients with near-perfect accuracy from unprotected updates. The authors evaluate lightweight defenses—clipping followed by Gaussian noise and ensemble FL—to mitigate this privacy risk while preserving model utility, reporting differential‑privacy budgets and empirical results across multiple attack classifiers and data partitions.
arXiv:2607. 01313v1 Announce Type: cross Abstract: In practice, most commercial LLM providers do not publicly release details of underlying LLM architectures.