The intelligent fault diagnosis model has made a significant development, whose high-precision results rely on a large amount of labeled data. However, in the actual industrial environment, it is very difficult to obtain a large amount of labeled data. It will make it difficult for the fault diagnosis model to converge with limited labeled industrial data. To address this paradox, we propose a novel unsupervised domain adaptation framework (M-Net) for fault diagnosis of rotating machinery, which only requires unlabeled industrial data. The M-Net will be pretrained using the labeled data, which can be accessed through the labs. In this stage, we propose a multi-scale feature extractor that can extract and fuse multi-scale features. This operation will generalize the features further. Then, we will align the distribution of the labeled data and unlabeled industrial data using the generator model based on multi-kernel maximum mean discrepancy. This will reduce the distribution distance between the labeled data and the unlabeled industrial data. For now, the unsupervised domain adaptation problem has shifted to a semi-supervised domain adaptation problem. The results, obtained through experimental comparison, demonstrate that the M-Net can achieve an accuracy of over 99.99% with labeled data and a maximum transfer accuracy of over 99% with unlabeled industrial data.

With the widespread adoption of biometric technology in consumer electronics (e.g., smart locks, wearables), biometric data transmission increasingly faces security threats like relay attacks, tampering, and spoofing. Traditional passive defenses are ineffective in anonymized traffic tracing, suffer from high latency, and lack adaptability. Network flow watermarking (NFW), an active traffic analysis technique, embeds covert identifiers to secure biometric data and trace attacks. However, existing solutions in consumer electronics still face challenges, including synchronization instability, detectability risks, and the trade-off between robustness and concealment. To address these challenges, we propose FRW-TRACE, a forensic-ready watermarking framework with enhanced traceability, which innovatively integrates temporal group statistical features and packet sequence features as a dual-modal embedding carrier. To mitigate synchronization loss due to temporal perturbations, we develop a packet sequence-based synchronization mechanism that ensures reliable extraction of synchronization information. Additionally, to alleviate watermark degradation in high packet loss environments, we design an interval packet counting-based resilient encoding strategy, which improves watermark robustness against network disruptions and effectively evades detection by the Kolmogorov-Smirnov (K-S) test. Extensive experimental evaluations demonstrate that FRW-TRACE achieves an average detection accuracy of 100% under high latency jitter conditions and maintains over 90% accuracy in high packet loss scenarios. Furthermore, comparative analysis shows that under identical interference conditions, FRW-TRACE improves detection accuracy by 2%-4% compared to the state-of-the-art NFW methods.