Investigação da Importância de Atributos de Rede para Detecção de Ataques Através do Protocolo MQTT na IoMT
Resumo
A crescente utilização de dispositivos IoMT em ambientes de saúde exige mecanismos eficientes e interpretáveis de detecção de intrusões. Este trabalho analisa a importância dos atributos de rede na detecção de seis classes de ataques MQTT do CICIoMT2024, utilizando XGBoost, MLP e Naive Bayes. A técnica SHAP identificou os atributos mais relevantes, posteriormente removidos em um experimento de ablação. O XGBoost apresentou o melhor desempenho, com F1-Macro de 0,9923 ± 0,0008, e o atributo IAT apresentou a maior importância SHAP global (0,1657). A análise por classe evidenciou diferentes impactos da remoção dos atributos.Referências
Dadkhah, S., Neto, E. C. P., Ferreira, R., Molokwu, R. C., Sadeghi, S., and Ghorbani, A. A. (2024). CICIoMT2024: A benchmark dataset for multi-protocol security assessment in IoMT. Internet of Things, 28:101351.
ElRushaidat, D., Sammak, T., Ghannam, Y., and Alkhalil, B. (2026). Hierarchical versus flat machine learning model for intrusion detection in secure IoT healthcare environment. Discover Internet of Things, 6(1):154.
Kharoubi, K., Cherbal, S., Mechta, D., and Gawanmeh, A. (2025). Network Intrusion Detection System Using Convolutional Neural Networks: NIDS-DL-CNN for IoT Security. Cluster Computing, 28(4):219.
Kumar, R. and Swarnkar, M. (2026). MGIDS: A Graph-based Intrusion Detection System for MQTT-based IoT Networks. ACM Trans. Internet Things, 7(4).
Lui, P., Siqueira, L., Kazienko, J., Quincozes, V., Quincozes, S., and Welfer, D. (2025). On the Performance of Cyber-Biomedical Features for Intrusion Detection in Healthcare 5.0. In Anais do XXV Simpósio Brasileiro de Computação Aplicada à Saúde, pages 389–400, Porto Alegre, RS, Brasil. SBC.
Nomachi, R. Y. G. and Giannotti, M. A. (2025). Explainable AI and causal inference to assess the impact of São Paulo’s metro stations. Master’s thesis, USP.
Quincozes, V. E., Quincozes, S. E., Kazienko, J. F., Gama, S., Cheikhrouhou, O., and Koubaa, A. (2024). A survey on IoT application layer protocols, security challenges, and the role of explainable AI in IoT (XAIoT). International Journal of Information Security, 23(3):1975–2002.
Salehpour, A., Balafar, M. A., and Souri, A. (2025). An optimized intrusion detection system for resource-constrained IoMT environments: enhancing security through efficient feature selection and classification. The Journal of Supercomputing, 81(6):783.
Turgut, Z. and Başarslan, M. S. (2025). XBiDeep: A novel explainable artificial intelligence based intrusion detection system for Internet of Medical Things environment. Internet of Things, 33:101675.
Veledar, E., Zhou, L., Veledar, O., Gardener, H., Gutierrez, C. M., Romano, J. G., and Rundek, T. (2025). Synthesizing Explainability Across Multiple ML Models for Structured Data. Algorithms, 18(6).
ElRushaidat, D., Sammak, T., Ghannam, Y., and Alkhalil, B. (2026). Hierarchical versus flat machine learning model for intrusion detection in secure IoT healthcare environment. Discover Internet of Things, 6(1):154.
Kharoubi, K., Cherbal, S., Mechta, D., and Gawanmeh, A. (2025). Network Intrusion Detection System Using Convolutional Neural Networks: NIDS-DL-CNN for IoT Security. Cluster Computing, 28(4):219.
Kumar, R. and Swarnkar, M. (2026). MGIDS: A Graph-based Intrusion Detection System for MQTT-based IoT Networks. ACM Trans. Internet Things, 7(4).
Lui, P., Siqueira, L., Kazienko, J., Quincozes, V., Quincozes, S., and Welfer, D. (2025). On the Performance of Cyber-Biomedical Features for Intrusion Detection in Healthcare 5.0. In Anais do XXV Simpósio Brasileiro de Computação Aplicada à Saúde, pages 389–400, Porto Alegre, RS, Brasil. SBC.
Nomachi, R. Y. G. and Giannotti, M. A. (2025). Explainable AI and causal inference to assess the impact of São Paulo’s metro stations. Master’s thesis, USP.
Quincozes, V. E., Quincozes, S. E., Kazienko, J. F., Gama, S., Cheikhrouhou, O., and Koubaa, A. (2024). A survey on IoT application layer protocols, security challenges, and the role of explainable AI in IoT (XAIoT). International Journal of Information Security, 23(3):1975–2002.
Salehpour, A., Balafar, M. A., and Souri, A. (2025). An optimized intrusion detection system for resource-constrained IoMT environments: enhancing security through efficient feature selection and classification. The Journal of Supercomputing, 81(6):783.
Turgut, Z. and Başarslan, M. S. (2025). XBiDeep: A novel explainable artificial intelligence based intrusion detection system for Internet of Medical Things environment. Internet of Things, 33:101675.
Veledar, E., Zhou, L., Veledar, O., Gardener, H., Gutierrez, C. M., Romano, J. G., and Rundek, T. (2025). Synthesizing Explainability Across Multiple ML Models for Structured Data. Algorithms, 18(6).
Publicado
21/10/2026
Como Citar
LIMA, Carlos E. C. de; LUI, Pedro H.; SIQUEIRA, Lucas P.; KAZIENKO, Juliano F.; QUINCOZES, Vagner E.; QUINCOZES, Silvio E.; WELFER, Daniel.
Investigação da Importância de Atributos de Rede para Detecção de Ataques Através do Protocolo MQTT na IoMT. In: ESCOLA REGIONAL DE REDES DE COMPUTADORES (ERRC), 23. , 2026, Santa Maria/RS.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2026
.
p. 103-108.
DOI: https://doi.org/10.5753/errc.2026.36173.
