Device identification using optimized digital footprints

The rapidly increasing number of internet things (IoT) and non-IoT devices has imposed new security challenges to network administrators. Accurate device identification in the increasingly complex structures is necessary. In this paper, a fingerprinting (DFP) method been proposed for identification, based on digital footprints, which use communication over network. A subset nine features have selected from transport layers single transmission control protocol/internet protocol packet attribute evaluators Weka, generate device-specific signatures. evaluated two online datasets, an experimental dataset, using different supervised machine learning (ML) algorithms. Results shown that able distinguish type with up 100% precision random forest (RF) classifier, classify individual 95.7% precision. These results demonstrate applicability DFP order provide more secure robust