Skip to Main content Skip to Navigation
Conference papers

Autonomous Identification of IoT Device Types based on a Supervised Classification

Abstract : A wide diversity of IoT devices are connected in various smart homes with an extremely rapid growth. Identifying IoT devices as they connect to the network enables better devices and services management. However, autonomous identification is very challenging in such heterogeneous environments. In this paper, we present a near-real time classification approach, based on network features that are extracted both from characteristics of traffic flows and from the payloads of the packets, to discriminate device types. Our solution automatically identifies a newly connected device to the home network. Furthermore, we evaluate the performance of our method using a representative and heterogeneous set of real IoT devices. Our results show autonomous recognition with 97% average accuracy, based on decision tree models using the one-vs.-all method.
Document type :
Conference papers
Complete list of metadata

https://hal.archives-ouvertes.fr/hal-02913093
Contributor : Ludovic Noirie <>
Submitted on : Friday, August 7, 2020 - 2:42:37 PM
Last modification on : Tuesday, March 23, 2021 - 9:28:02 AM

Identifiers

Citation

Nesrine Ammar, Ludovic Noirie, Sébastien Tixeuil. Autonomous Identification of IoT Device Types based on a Supervised Classification. ICC 2020 - 2020 IEEE International Conference on Communications (ICC), Jun 2020, Virtual conference, Ireland. ⟨10.1109/ICC40277.2020.9148821⟩. ⟨hal-02913093⟩

Share

Metrics

Record views

80