Cybersecure Intelligent Sensor Framework for Smart Buildings: AI-Based Intrusion Detection and Resilience Against IoT Attacks
Preprint
- 15 September 2025
- preprint
- Published by MDPI AG in Preprints.org
Abstract
Due to the fast development of the Internet of Things (IoT) in smart buildings, the efficiency of operations, personal comfort, and sustainable operations have all been enhanced. But with all this dependence on potentially linked systems comes vital cybersecurity weaknesses. When such weaknesses are used to attack, they may lead to the compromise of the security of both an individual device as well as a building infrastructure. The present paper introduces a new cybersecure intelligent sensor framework that will be able to protect smart buildings against a wide variety of IOT related cyberattacks. The key element within this structure is a sophisticated AI automated intrusion detector (IDS) that can recognize, categorize and eliminate prospective threats in an instant using machine learning routines. The system uses the combination of intelligent sensor networks with AI-based analytics to continuously observe the environment and system data, with anomalous behaviours being indicators of a security breach being detected. The combination of predictive modelling and automated threat responses will allow the proposed system to achieve resilience against many attacks, including but not limited to denial of service, unauthorized access, and data manipulation. Widespread simulation and testing have shown that the system has a high detection rate, low false alarms, and a fast response time to help secure infrastructure buildings in smart buildings whilst Downtime is minimal. The results demonstrate the possible future of AI-enhanced cybersecurity systems in developing the IoT-based smart building security and enjoyment.Keywords
All Related Versions
- Version 1, 2025-09-15, Preprints
- Published version: Sensors, 25 (24), 7680.
This publication has 19 references indexed in Scilit:
- An end-to-end intrusion detection system with IoT dataset using deep learning with unsupervised feature extractionInternational Journal of Information Security, 2024
- Autonomous Federated Learning for Distributed Intrusion Detection Systems in Public NetworksIEEE Access, 2023
- Network Intrusion Detection via Flow-to-Image Conversion and Vision Transformer ClassificationIEEE Access, 2022
- Recurrent deep learning-based feature fusion ensemble meta-classifier approach for intelligent network intrusion detection systemComputers and Electrical Engineering, 2022
- Hardware-Assisted Machine Learning in Resource-Constrained IoT Environments for Security: Review and Future ProspectiveIEEE Access, 2022
- IoT Intrusion Detection System Using Deep Learning and Enhanced Transient Search OptimizationIEEE Access, 2021
- Unified Deep Learning approach for Efficient Intrusion Detection System using Integrated Spatial–Temporal FeaturesKnowledge-Based Systems, 2021
- IGAN-IDS: An imbalanced generative adversarial network towards intrusion detection system in ad-hoc networksAd Hoc Networks, 2020
- AI-IDS: Application of Deep Learning to Real-Time Web Intrusion DetectionIEEE Access, 2020
- Improving AdaBoost-based Intrusion Detection System (IDS) Performance on CIC IDS 2017 DatasetJournal of Physics: Conference Series, 2019