Detection of Anomalies in Internet of Things (IoT) Devices and Sensors

Though IoT devices and sensors are intended to make our lives easier, they can be vulnerable to malicious activities. As such, a critical part of IoT security is the development of algorithms that can detect anomalies in these devices and sensors. This study aimed to review the various approaches to anomaly detection in the context of IoT devices and sensors. The objectives are to evaluate unsupervised, supervised, reinforcement, and deep learning methods. The research methodology adopted for this study was explanatory. This study focuses on the detection of anomalies in IoT devices and sensors. This involves literature research using sources like web sources, in-print material, journals, textbooks, and articles. In conclusion, this study evaluated the various approaches to anomaly detection in the context of IoT devices and sensors. Methods such as unsupervised, supervised, reinforcement, and deep learning are critically reviewed.