Thuat Nguyen-Khanh

Thuat Nguyen-Khanh

Auto-generated profile, based on publicly available information

Thuat Nguyen-Khanh

University of Information Technology

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications

5 January 2026 in IEEE Access
Verified
Journal Article
Open access
BPBao Pham-Thai
Bao Pham-Thai
MNMy Nguyen-Le-Ha
My Nguyen-Le-Ha
TNThuat Nguyen-Khanh
Thuat Nguyen-Khanh
QLQuan Le-Trung
Quan Le-Trung
The rapid development of Internet-connected devices has driven the rise of smart cities, where large-scale Internet of Things (IoT) systems require efficient infrastructure management with centralized monitoring and control. In this context, Software-Defined Networking (SDN) architecture offers an effective solution to fulfill centralized management needs and optimize network quality. By decoupling the data plane from the control plane, SDN consolidates control over network traffic, allowing flexible configuration and real-time traffic optimization. However, the separation also exposes SDN to severe attacks, particularly Distributed Denial of Service (DDoS) attacks, along with other exploitation techniques targeting its architecture. In this study, we propose a network traffic classification model for SDN environments that leverages multiple Machine Learning (ML) and Deep Learning (DL) models integrated with Explainable Artificial Intelligence (XAI) technology. This approach enables precise and transparent model decisions while reducing computational costs and improving classification accuracy. The proposed model achieves up to 99% accuracy across various network attack detection and classification scenarios within SDN, both with and without XAI assistance. Notably, the incorporation of XAI allows the model to retain high performance using only approximately 15% of the original feature set, significantly outperforming models without XAI. This feature reduction contributes to substantial decreases in training time, inference time, and model size, demonstrating the method’s feasibility and effectiveness for real-time deployment in SDN-based networks.


23 February 2024
Verified
Conference Paper
Open access
TNThuat Nguyen-Khanh
Thuat Nguyen-Khanh
BNBao Nguyen-Van
Bao Nguyen-Van
PPPhat Phan-Trung
Phat Phan-Trung
LVLuan Van-Thien
Luan Van-Thien
CDChuong Dang-Le-Bao
Chuong Dang-Le-Bao
QLQuan Le-Trung
Quan Le-Trung
Weather forecasting is an essential task in smart agriculture, wide-area weather reports typically predict over a large geographic area likes city or province. In this paper, we deploy an environmental monitoring station on the rooftop of the building E at the University of Information Technology (UIT). In addition, we build a system to collect these environmental data and store them in a dataset, namely UiTiOt, whose metrics are temperature, humidity, pressure, nitro dioxide, etc. Noisy data points are filtered out using Kalman filter before training with Deep learning models such as Autoregressive integrated moving average (ARIMA), Gated Recurrent Unit (GRU), Long Short Term Memory (LSTM), and Bidirectional LSTM (Bi-LSTM). The process is evaluated using various metrics such as mean absolute error (MAE), mean squared error (MSE), root mean squared error (RMSE), root relative squared error (RRSE), and mean absolute percentage error (MAPE). The results show that ARIMA is the best model among the four models when measuring on three datasets. Furthermore, we observe that the smoothed UiTiOt dataset achieves lower values in measured metric indices than the unfiltered one.


11 November 2025 in EAI Endorsed Transactions on Industrial Networks and Intelligent Systems
Verified
Journal Article
Open access
TNThuat Nguyen-Khanh
Thuat Nguyen-Khanh
APAnh Pham-Nguyen-Hai
Anh Pham-Nguyen-Hai
LVLuan Van-Thien
Luan Van-Thien
QLQuan Le-Trung
Quan Le-Trung
The Internet of Things (IoT) has emerged as a foundational platform for driving intelligent solutions, playing a central role in the Fourth Industrial Revolution. Its potential lies in enabling seamless connectivity and real-time data exchange among diverse devices and systems, thereby powering advanced applications such as intelligent transportation, smart healthcare, precision agriculture, and automated manufacturing. These solutions promise to improve efficiency, optimize resource utilization, and enhance decision-making across various sectors. However, this potential is challenged by some issues, including security vulnerabilities, privacy concerns, and significant heterogeneity arising from the vast diversity of devices, communication protocols, and data formats. In this paper, we develop a multimodal deep learning solution to detect DDoS attacks on IoT infrastructure based on two data types: packet-based data and flow-based data. Firstly, the datasets containing packets labeled as benign or attack are processed into two branches: packet-based and flow-based features. Then, each branch is trained using two independent CNN models. Finally, the feature information extracted from both modalities is fused and fed into a concatenation-based classifier for DDoS attack detection. Experimental results on Edge-IIoTset and CiCIoMT2024 datasets indicate that the multimodal deep learning model within a decentralized machine learning architecture achieves performance comparable to centralized machine learning. In addition, our proposal is also robust to non-independent and identically distributed (non-IID) data in decentralized machine learning architecture.