US 12,245,032 B2
Methods and apparatus to discriminate authentic wireless Internet-of-Things devices
Tao Zheng, Beijing (CN); Xiaoyu Wang, Beijing (CN); and Xin Wang, Beijing (CN)
Assigned to ORANGE, Issy-les-Moulineaux (FR)
Appl. No. 17/602,663
Filed by Orange, Issy-les-Moulineaux (FR)
PCT Filed Apr. 9, 2020, PCT No. PCT/IB2020/000356
§ 371(c)(1), (2) Date Oct. 8, 2021,
PCT Pub. No. WO2020/208426, PCT Pub. Date Oct. 15, 2020.
Claims priority of application No. PCT/CN2019/081970 (WO), filed on Apr. 9, 2019.
Prior Publication US 2022/0182824 A1, Jun. 9, 2022
Int. Cl. H04W 12/06 (2021.01); G06N 3/044 (2023.01)
CPC H04W 12/06 (2013.01) [G06N 3/044 (2023.01)] 15 Claims
OG exemplary drawing
 
1. An apparatus configured to discriminate authentic wireless-IoT-devices, the apparatus comprising:
a receiver configured to receive wireless data from IoT devices;
a trained machine-learning module configured to receive and analyse the wireless data received by the receiver; and
an interface to output the result of analysis of the wireless data by the trained machine-learning module as an indication of identification of a wireless IoT device as one of an authentic wireless IoT device or an inauthentic wireless IoT device, wherein an inauthentic wireless IoT device is at least one of modified, fake, illegal, or unauthorized, wherein the trained machine-learning module is arranged to analyse data from frame headers of frames of the wireless data received from the wireless IoT devices, wherein the trained machine-learning module is configured to analyse frame header data that excludes address data representing an address of a device providing the wireless data, wherein the trained machine-learning module is configured to analyse a set of parameters of the frame header data, a number of the parameters in the set of parameters being dependent upon an amount of difference between authentic and inauthentic wireless IoT devices.