Abstract:
Smart buildings, enabled by the widespread use of Internet of Things (IoT) devices, are becoming increasingly prevalent. However, this rise in IoT adoption also brings new security challenges, with smart buildings more vulnerable to cyber-attacks, including distributed denial-of-service (DDoS) attacks. DDoS attacks can cause significant damage to the building's network infrastructure, leading to financial losses and downtime.
This thesis proposes a machine learning (ML) based approach to detect DDoS attacks in smart buildings. The proposed solution employs various ML algorithms, including SVM, decision trees, Neural Network using TensorFlow and linear regression. These models are trained to analyse network traffic data collected from smart building devices and detect and classify network traffic patterns that indicate DDoS attacks.
To train the ML models, network traffic data collected from smart buildings is pre-processed to extract relevant features. The performance of the models is evaluated based on accuracy, precision, and recall metrics. The results show that the proposed ML-based approach outperforms traditional rule-based methods.
The proposed solution contributes to the development of efficient and effective cybersecurity mechanisms for smart buildings, enhancing their security and resilience against cyber threats. The generic nature of the proposed approach means that it can be applied to various types of smart buildings, making it a versatile solution for improving smart building cybersecurity.
In conclusion, this thesis demonstrates the effectiveness of ML techniques in detecting DDoS attacks in smart buildings. The proposed solution can be used to build secure and resilient smart.
buildings, ensuring the safety and privacy of occupants and the efficient operation of the building. This thesis adds to the growing body of research on improving the security of smart buildings, which is becoming increasingly important as smart building technology becomes more widespread.
