Anomaly Detection for IoT-Enabled Kitchen Area Network Using Machine Learning

IoT has eased our life by providing various smart applications such as Smart Homes, Industrial Automation, Smart Healthcare, Smart Traffic Monitoring, and Fleet Management, to name a few. Due to various vulnerabilities in IoT infrastructure and software, Anomaly Detection is one of the major concerns for application developers. In this paper, Smart Home application of the IoT paradigm has been presented. We proposed an Anomaly Detection System (ADS) for the Kitchen Area Network (KAN), which has been deployed with the help of the MQTT Broker and various sensors on the IoT Flock Emulator. IoT Network has been created with the help of an IoT Network Emulator. Network Traffic packet sniffer (i.e., Wireshark) is used to capture the flows. The KNN machine learning model is used for binary classification to detect anomalies. The proposed model achieves accuracy, recall, precision, and F-1 score of 94.37%, 94.31%, 95.40%, and 94.85%, respectively.