A METHOD FOR FEATURE SPACE CONSTRUCTION AND EXPANSION FOR INFORMATION SECURITY MONITORING OF IoT DEVICES BASED ON NETWORK TRAFFIC TIME SERIES SEGMENTATION
https://doi.org/10.53360/2788-7995-2026-3(23)-27
Abstract
Subject of the study. The widespread adoption of the Internet of Things (IoT) in smart homes, industrial automation, transportation, healthcare, and smart city infrastructures has led to a significant increase in the number of potential attack surfaces and unauthorized access points. Unlike conventional computing systems, IoT devices are characterized by limited hardware resources, including restricted memory capacity, low computational performance, limited communication bandwidth, and low power consumption. These constraints hinder the direct application of computationally intensive and cryptographic information security monitoring techniques without adaptation to resource-constrained environments. Method. To construct an informative feature representation of IoT device security states, a method based on network traffic time series segmentation using statistical, frequency-domain, and threshold-based criteria is proposed. The feature space is further expanded by incorporating external and internal factors affecting the observed feature values. Security state classification is performed using six machine learning algorithms of varying computational complexity: Logistic Regression, Naive Bayes, Support Vector Machine (SVM), Random Forest, Multilayer Perceptron (MLP), and Convolutional Neural Network (CNN). Main results. The proposed method was experimentally evaluated using both a laboratory testbed based on IoT nodes (IEEE 802.15.4, TelosB, ESP32) and the UNSW-NB15 dataset. The results demonstrate that network traffic time series segmentation combined with contextual information enables the construction of a more compact and discriminative feature space while reducing the computational cost of classifier training. Practical significance. The proposed approach can be applied to the development of lightweight information security monitoring systems for resource-constrained IoT infrastructures. Lightweight machine learning models can be deployed on edge devices, whereas more computationally intensive models can operate on gateways or cloud platforms, thereby reducing false alarm rates and improving the timeliness of cyberattack detection.
About the Authors
N. BazhayevKazakhstan
Nurzhan Bazhayev – postdoctoral fellow
2 Satbayev Street, Astana, 010000, Republic of Kazakhstan
A. Tokkuliyeva
Kazakhstan
Aizhan Tokkuliyeva – doctoral student
2 Satbayev Street, Astana, 010000, Republic of Kazakhstan
References
1. Lee J. A Cyber-Physical Systems Architecture for Industry 4.0 / J. Lee, B. Bagheri, H.A. Kao // Manufacturing Letters. – 2015. – Vol. 3. – P. 18-23.
2. Atzori L. The Internet of Things: A Survey / L. Atzori, A. Iera, G. Morabito // Computer Networks. – 2010. – Vol. 54, № 15. – P. 2787-2805.
3. Threat analysis of IoT networks using artificial neural network intrusion detection system / E. Hodo et al // 2016 International Symposium on Networks, Computers and Communications (ISNCC), Yasmine Hammamet, Tunisia, 2016. – Р. 1-6, https://doi.org/10.1109/ISNCC.2016.7746067.
4. Deep Learning for Cyber Security Intrusion Detection: Approaches, Datasets and Comparative Study / M.A. Ferrag et al // Journal of Information Security and Applications. – 2020. – Vol. 50.
5. Survey of intrusion detection systems: techniques, datasets and challenges / A. Khraisat et al // Cybersecur 2, 20 (2019). https://doi.org/10.1186/s42400-019-0038-7.
6. Alrawashdeh K. Toward an Online Anomaly Intrusion Detection System Based on Deep Learning / K. Alrawashdeh, C. Purdy // 2016 15th IEEE International Conference on Machine Learning and Applications (ICMLA), Anaheim, CA, USA, 2016. – Р. 195-200. https://doi.org/10.1109/ICMLA.2016.0040.
7. Sukhoparov M.E. Improving the Quality of the Identification of the Information Security State Based on Sample Segmentation / M.E. Sukhoparov, I.S. Lebedev // Aut. Control Comp. Sci. – 2023. – № 57. – Р. 1071-1075. https://doi.org/10.3103/S0146411623080321.
8. Semenov V.V. Identification of Abnormal Functioning of Devices of Cyber-Physical Systems / M.E. Sukhoparov, I.S. Lebedev // Internet of Things, Smart Spaces, and Next Generation Networks and Systems. NEW2AN ruSMART 2020. Lecture Notes in Computer Science. – 2020. – vol 12525. Springer, Cham. https://doi.org/10.1007/978-3-030-65726-0_1.
9. Segmenting Time Series: A Survey and Novel Approach / E. Keogh et al // Data Mining in Time Series Databases. – World Scientific, 2004.
10. Moustafa N. UNSW-NB15: a comprehensive data set for network intrusion detection systems (UNSW-NB15 network data set) / N. Moustafa, J. Slay // 2015 Military Communications and Information Systems Conference (MilCIS), Canberra, ACT, Australia. – 2015. – Р. 1-6. https://doi.org/10.1109/MilCIS.2015.7348942.
11. Fawcett T. An Introduction to ROC Analysis / T. Fawcett // Pattern Recognition Letters. – 2006. – Vol. 27, № 8. – P. 861-874.
Review
For citations:
Bazhayev N., Tokkuliyeva A. A METHOD FOR FEATURE SPACE CONSTRUCTION AND EXPANSION FOR INFORMATION SECURITY MONITORING OF IoT DEVICES BASED ON NETWORK TRAFFIC TIME SERIES SEGMENTATION. Bulletin of Shakarim University. Technical Sciences. 2026;(3(23)):278-288. (In Kazakh) https://doi.org/10.53360/2788-7995-2026-3(23)-27
JATS XML















