Browsing by Author "Sanjay, M."

Filter results by typing the first few letters
Now showing 1 - 2 of 2
  • No Thumbnail Available
    Item
    Prediction of Remaining Useful Life in MEMS Devices Using a Stacking Model Approach
    (Institute of Electrical and Electronics Engineers Inc., 2025) Sanjay, M.; Kumar, R.S.
    Accurate estimation of the Remaining Useful Life (RUL) of Micro-Electro-Mechanical Systems (MEMS) devices is essential for enhancing predictive maintenance in industrial settings. This paper proposes a robust stacking ensemble model for Remaining Useful Life (RUL) prediction of MEMS devices, integrating sensor and mechanical component data. Publicly available RUL datasets were augmented using Generative Adversarial Networks (GANs), ensuring diverse input data. Key algorithms-Bayesian Ridge, Random Forest, and XGBoost-were identified using Lazy Predict and combined with a Support Vector Regressor (SVR) as the meta-learner. Hyperparameter optimization was performed using the Deep Deterministic Policy Gradient (DDPG) algorithm, offering enhanced efficiency for large datasets and dynamic retraining. The model demonstrated superior performance based on RMSE and R2 metrics compared to traditional methods. Deployment was achieved via a Flask web application integrated with a CI/CD pipeline using GitHub Actions and Docker, ensuring scalability and reproducibility. This research contributes a scalable and efficient framework for advancing predictive maintenance in industrial automation. © 2025 IEEE.
  • No Thumbnail Available
    Item
    TCP SYN Flood Attack Detection Using Logistic Regression and Multi-Agent Reinforcement Learning
    (Institute of Electrical and Electronics Engineers Inc., 2025) Sanjay, M.; Arun Raj Kumar, P.
    In the realm of cybersecurity, Distributed Denial of Service (DDoS) attacks remain a continuous threat, particularly TCP SYN flood attacks due to their stealthiness and potential for disruption. In this paper, we propose a combination of Multi-Agent Reinforcement Learning (MARL) with logistic regression for enhancing TCP SYN attack detection, leveraging Actor-Critic as the reinforcement learning algorithm. A novel approach is introduced for hyperparameter optimization using MARL, offering an alternative to traditional techniques such as GridSearchCV and RandomSearchCV. We present a comparative analysis between traditional logistic regression and MARL enhanced approaches, evaluating their performance using metrics such as accuracy, false negatives, and false positives. Results demonstrate that our proposed approach significantly improves detection accuracy and reduces false positives, underscoring its potential in bolstering cybersecurity defenses against sophisticated DDoS threats. © 2025 IEEE.