Machine Learning for IoT Device Identification and Detection
Machine Learning for the Detection and Identification of Internet of Things (IoT) Devices: A Survey
We provide a comprehensive survey on machine learning technologies for the identification of devices and the detection of compromised or falsified ones from the viewpoint of passive surveillance agents or network operators.
We classify the device-specific pattern recognition, deep learning enabled device identification, unsupervised device identification, and abnormal device detection into four categories: device-specific pattern recognition, unsupervised device identification, device-specific pattern recognition, and abnormal device detection.
We also discuss various machine learning-related enabling technologies for this purpose.
These enabling technologies include learning algorithms, feature engineering on network traffic traces and wireless signals, continual learning, and abnormality detection.
Authors
Yongxin Liu, Jian Wang, Jianqiang Li, Shuteng Niu, Houbing Song