Cyber-Physical Systems integrated with information technologies introduce vulnerabilities that extend beyond traditional cyber threats. Attackers can non-invasively manipulate sensors and spoof controllers, which in turn increases the autonomy of the system. Even though the focus on protecting against sensor attacks increases, there is still uncertainty about the optimal timing for attack detection. Existing systems often struggle to manage the trade-off between latency and false alarm rate, leading to inefficiencies in real-time anomaly detection. This paper presents a framework designed to monitor, predict, and control dynamic systems with a particular emphasis on detecting and adapting to changes, including anomalies such as “drift” and “attack”. The proposed algorithm integrates a Transformer-based Attention Generative Adversarial Residual model, which combines the strengths of generative adversarial networks, residual networks, and attention algorithms. The system operates in two phases: offline and online. During the offline phase, the proposed model is trained to learn complex patterns, enabling robust anomaly detection. The online phase applies a trained model, where the drift adapter adjusts the model to handle data changes, and the attack detector identifies deviations by comparing predicted and actual values. Based on the output of the attack detector, the controller makes decisions then the actuator executes suitable actions. Finally, the experimental findings show that the proposed model balances detection accuracy of 99.25%, precision of 98.84%, sensitivity of 99.10%, specificity of 98.81%, and an F1-score of 98.96%, thus provides an effective solution for dynamic and safety-critical environments.

Biometric authentication provides a reliable, user-specific approach for identity verification, significantly enhancing access control and security against unauthorized intrusions in cybersecurity. Unimodal biometric systems that rely on either face or voice recognition encounter several challenges, including inconsistent data quality, environmental noise, and susceptibility to spoofing attacks. To address these limitations, this research introduces a robust multi-modal biometric recognition framework, namely Quantum-Enhanced Biometric Fusion Network. The proposed model strengthens security and boosts recognition accuracy through the fusion of facial and voice features. Furthermore, the model employs advanced pre-processing techniques to generate high-quality facial images and voice recordings, enabling more efficient face and voice recognition. Augmentation techniques are deployed to enhance model performance by enriching the training dataset with diverse and representative samples. The local features are extracted using advanced neural methods, while the voice features are extracted using a Pyramid-1D Wavelet Convolutional Bidirectional Network, which effectively captures speech dynamics. The Quantum Residual Network encodes facial features into quantum states, enabling powerful quantum-enhanced representations. These normalized feature sets are fused using an early fusion strategy that preserves complementary spatial-temporal characteristics. The experimental validation is conducted using a biometric audio and video dataset, with comprehensive evaluations including ablation and statistical analyses. The experimental analyses ensure that the proposed model attains superior performance, outperforming existing biometric methods with an average accuracy of 98.99%. The proposed model improves recognition robustness, making it an efficient multimodal solution for cybersecurity applications.