To search, Click below search items. |
| All Published Papers Search Service |
|
|
|
Title |
Enhancing Internet of Things Security with Random Forest-Based Anomaly Detection |
|
Author |
Ahmed Al Shihimi, Muhammad R Ahmed, Thirein Myo, and Badar Al Baroomi |
| Citation |
Vol. 24 No. 6 pp. 67-76 |
|
Abstract |
The Internet of Things (IoT) has revolutionized communication and device operation, but it has also brought significant security challenges. IoT networks are structured into four levels: devices, networks, applications, and services, each with specific security considerations. Personal Area Networks (PANs), Local Area Networks (LANs), and Wide Area Networks (WANs) are the three types of IoT networks, each with unique security requirements. Communication protocols such as Wi-Fi and Bluetooth, commonly used in IoT networks, are susceptible to vulnerabilities and require additional security measures. Apart from physical security, authentication, encryption, software vulnerabilities, DoS attacks, data privacy, and supply chain security pose significant challenges. Ensuring the security of IoT devices and the data they exchange is crucial. This paper utilizes the Random Forest Algorithm from machine learning to detect anomalous data in IoT devices. The dataset consists of environmental data (temperature and humidity) collected from IoT sensors in Oman. The Random Forest Algorithm is implemented and trained using Python, and the accuracy and results of the model are discussed, demonstrating the effectiveness of Random Forest for detecting IoT device data anomalies. |
|
Keywords |
Internet of Things, Challenges, Security, Random Forest |
|
URL |
