Prototipe AI-IoT Edge Berbasis Raspberry Pi dan TinyML untuk Pemantauan Jaringan Kampus secara Real-Time
DOI:
https://doi.org/10.32492/jeetech.v7i1.7112Keywords:
Network monitoring, Edge computing, Autoencoder, TinyMLAbstract
Complex campus networks featuring server-based services and the growing Internet of Things (IoT) require near-real-time monitoring systems without incurring significant overhead. This study proposes a lightweight Artificial Intelligence-Internet of Things (AI-IoT)-based network monitoring prototype on an edge computing platform, utilizing an unsupervised autoencoder for anomaly detection. This prototype is implemented out-of-band on a Raspberry Pi 4 Model B device that serves as both a collection and inference node. The deep learning model on the TensorFlow Lite framework is compressed using TinyML for compatibility with small devices. The results use a dataset of 600,000 labeled flows that illustrate the trade-off in operational flexibility. At the P70 threshold, an F1-Score of 0.60 (precision 0.96, recall 0.43) is obtained, and in the P95 scenario, false positives can be completely eliminated. The edge infrastructure demonstrated excellent performance with an average batch processing latency of 74 ms and a throughput of over 300 flows/second with a constant Random Access Memory (RAM) usage of 2.8%.
References
J. Li, N. B. Linsangan, and H. Dong, “Campus Network Traffic Prediction and Anomaly Detection Based on Deep Learning,” Int. J. Emerg. Technol. Adv. Appl., vol. 1, no. 7, pp. 8–13, Aug. 2024, doi: 10.62677/IJETAA.2407123.
R. Gutierrez, W. Villegas-Ch, and J. Govea, “Modular middleware for IoT: scalability, interoperability and energy efficiency in smart campus,” Front. Commun. Netw., vol. 6, p. 1672617, Sep. 2025, doi: 10.3389/frcmn.2025.1672617.
A. Fathima and G. S. Devi, “Enhancing university network management and security: a real-time monitoring, visualization & cyber attack detection approach using Paessler PRTG and Sophos Firewall,” Int. J. Syst. Assur. Eng. Manag., Aug. 2024, doi: 10.1007/s13198-024-02448-y.
R. M. Oviedo, F. Ramos, S. Gormus, P. Kulkarni, and M. Sooriyabandara, “A Comparison of Centralized and Distributed Monitoring Architectures in the Smart Grid,” IEEE Syst. J., vol. 7, no. 4, pp. 832–844, Dec. 2013, doi: 10.1109/JSYST.2013.2246033.
Xiaojiang Du, “Toward efficient distributed network monitoring,” in IEEE International Conference on Performance, Computing, and Communications, 2004, Phoenix, AZ, USA: IEEE, 2004, pp. 87–94. doi: 10.1109/PCCC.2004.1394950.
F. Zhou, M. Yuan, Y. Liu, H. Zhang, M. Gu, and T. Zhou, “Niect: A Model for Intrusion Security Detection Applied to Campus Video Surveillance Edge Networks,” in 2024 IEEE 11th International Conference on Cyber Security and Cloud Computing (CSCloud), Shanghai, China: IEEE, Jun. 2024, pp. 24–29. doi: 10.1109/CSCloud62866.2024.00012.
S. Hussain et al., “Edge AI-based self-learning technique for mitigating DDoS attacks in WSN,” Comput. Netw., vol. 273, p. 111769, Dec. 2025, doi: 10.1016/j.comnet.2025.111769.
P. V. Sithole and T. Justice Lavhengwa, “Exploring theories towards deploying edge computing in South African Higher Education Institutions,” in 2025 Conference on Information Communications Technology and Society (ICTAS), Durban, South Africa: IEEE, Jul. 2025, pp. 1–5. doi: 10.1109/ICTAS64866.2025.11155306.
S. Heydari and Q. H. Mahmoud, “Tiny Machine Learning and On-Device Inference: A Survey of Applications, Challenges, and Future Directions,” Sensors, vol. 25, no. 10, p. 3191, May 2025, doi: 10.3390/s25103191.
H.-A. Rashid, U. Kallakuri, and T. Mohsenin, “TinyM2 Net-V2: A Compact Low-power Software Hardware Architecture for M ulti m odal Deep Neural Networks,” ACM Trans. Embed. Comput. Syst., vol. 23, no. 3, pp. 1–23, May 2024, doi: 10.1145/3595633.
J. Leslin, M. Trapp, and M. Andraud, “Hardware-efficient tractable probabilistic inference for TinyML Neurosymbolic AI applications,” in 2025 IEEE International Conference on Omni-layer Intelligent Systems (COINS), Madison, WI, USA: IEEE, Aug. 2025, pp. 1–6. doi: 10.1109/COINS65080.2025.11125733.
J. D. Velasquez, L. Cadavid, and C. J. Franco, “Emerging trends and strategic opportunities in tiny machine learning: A comprehensive thematic analysis,” Neurocomputing, vol. 648, p. 130746, Oct. 2025, doi: 10.1016/j.neucom.2025.130746.
A. Vikram and Mohana, “Anomaly detection in Network Traffic Using Unsupervised Machine learning Approach,” in 2020 5th International Conference on Communication and Electronics Systems (ICCES), Coimbatore, India: IEEE, Jun. 2020, pp. 476–479. doi: 10.1109/ICCES48766.2020.9137987.
Dj. K. Nkashama et al., “ANADOE: Auto-Encoder-Based Network Anomaly Detection with Outlier Exposure,” Mar. 12, 2025. doi: 10.36227/techrxiv.174181618.89242519/v1.
L. A. K. Mekemte and G. Chalhoub, “On the Use of Autoencoders in Unsupervised Learning for Intrusion Detection Systems,” in Ubiquitous Networking, vol. 14757, O. Habachi, G. Chalhoub, H. Elbiaze, and E. Sabir, Eds., in Lecture Notes in Computer Science, vol. 14757. , Cham: Springer Nature Switzerland, 2024, pp. 54–69. doi: 10.1007/978-3-031-62488-9_5.
F. S. Alrayes, M. Zakariah, S. U. Amin, Z. Iqbal Khan, and M. Helal, “Intrusion Detection in IoT Systems Using Denoising Autoencoder,” IEEE Access, vol. 12, pp. 122401–122425, 2024, doi: 10.1109/ACCESS.2024.3451726.
M. Jagarajan and R. Jayaraman, “IoT edge computing and deep learning analytics: A survey,” presented at the 4TH INTERNATIONAL CONFERENCE ON INTERNET OF THINGS 2023: ICIoT2023, Kattankalathur, India, 2024, p. 020283. doi: 10.1063/5.0217187.
M. A. Ali and F. Dornaika, “Edge Artificial Intelligence: A Systematic Review of Evolution, Taxonomic Frameworks, and Future Horizons,” Oct. 01, 2025, arXiv: arXiv:2510.01439. doi: 10.48550/arXiv.2510.01439.
T. K. S. Flores, D. G. Costa, and I. Silva, “TensorFlores: An enhanced Python-based TinyML framework,” SoftwareX, vol. 31, p. 102224, Sep. 2025, doi: 10.1016/j.softx.2025.102224.
S. Wei, “Edge-Based Real-Time IIoT Anomaly Detection Using Semi-Supervised CNN-Attention Model with Cross-Protocol Capabilities,” Informatica, vol. 49, no. 4, Dec. 2025, doi: 10.31449/inf.v49i4.10508.
M. Y. Jo and H. J. Kim, “A Comparative Study of Lightweight, Sparse Autoencoder-Based Classifiers for Edge Network Devices: An Efficiency Analysis of Feed-Forward and Deep Neural Networks,” Sensors, vol. 25, no. 20, p. 6439, Oct. 2025, doi: 10.3390/s25206439.
S. Jeon, C. Park, G. Lee, S. Kim, and B. Gu, “Threshold Determination Method in Anomaly Detection using LSTM Autoencoder,” J. Korean Inst. Inf. Technol., vol. 21, no. 4, pp. 21–30, Apr. 2023, doi: 10.14801/jkiit.2023.21.4.21.
H. Zhang and T. Cao, “A Hybrid Approach to Network Intrusion Detection Based On Graph Neural Networks and Transformer Architectures,” in 2024 14th International Conference on Information Science and Technology (ICIST), Chengdu, China: IEEE, Dec. 2024, pp. 574–582. doi: 10.1109/ICIST63249.2024.10805457.
Azyk Orozonova, “Ai-Driven Anomaly Detection in Iot Networks Using Advanced Machine Learning Techniques,” Int. J. Appl. Math., vol. 38, no. 11s, pp. 1657–1671, Nov. 2025, doi: 10.12732/ijam.v38i11s.1277.
L. Van Langendonck, I. Castell-Uroz, and P. Barlet-Ros, “PPT-GNN: A Practical Pretrained Spatio-Temporal Graph Neural Network for Network Security,” in 2025 IEEE European Symposium on Security and Privacy Workshops (EuroS&PW), Venice, Italy: IEEE, Jun. 2025, pp. 169–175. doi: 10.1109/EuroSPW67616.2025.00026.
S. Jamshidi, F. Erfan, O. Abdul-Wahab, M. Bellaiche, and F. Khomh, “Lightweight Autoencoder-Isolation Forest Anomaly Detection for Green IoT Edge Gateways,” 2025, arXiv. doi: 10.48550/ARXIV.2511.18235.
M. B. Musthafa, S. Huda, T. T. Nguyen, Y. Kodera, and Y. Nogami, “Optimized Ensemble Deep Learning for Real-Time Intrusion Detection on Resource-Constrained Raspberry Pi Devices,” IEEE Access, vol. 13, pp. 113544–113556, 2025, doi: 10.1109/ACCESS.2025.3584373.
Published
How to Cite
Issue
Section
Copyright (c) 2026 Bima Aulia Firmandani, F Yudi Limpraptono, Michael Ardhita, Machrus Ali
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
