Copyright: ThalesSource : Thales

Challenge

AI x IoT

ApplyAppliquer
This challenge is over, applications are closed.

Challenge Overview

Description du challenge

Guavus (a Thales company) is offering service providers, enterprises and governments the solutions to analyze data the instant it’s captured, hence providing the unique ability to be decisive in today’s data-driven world.Today, Network Operators (XNO) are facing the Connected IoT Devices business opportunity, which will be amplified with the availability of 5G. To embrace this business, XNOs must be ready to provide Security and Maintenance services to their enterprise customers.

The exponential amount of connected IoT devices makes it hard to track each device’s traffic.The challenge is to learn the IoT device's normal behavior in order to identify when it deviates from its baseline considering the vast amount of network records and the ongoing addition of new devices attached to the network.

In particular, the challenge implies looking at the layers of the network traffic to learn the normal behavior of an IoT device and classifying it with peer devices. The classification must consider enough attributes so that when a device deviates from its normal behavior it could be identified. The solution design must be adapted to the vast amount of network traffic.

Copyright: 945_7039.NEF
Copyright: 945_7039.NEF
Copyright: Thales
Copyright: Thales

Wanted

We are looking for AI-first solutions that will contribute to solving this challenge by:

  • Automatically learning existing a newly connected devices’ normal behavior
  • Apply an anomaly detection algorithm to trigger notification
  • Report on the IoT activity at Enterprise level

One of the major challenge being to apply AI at the scale of a large Mobile Network Operator.

Projects you could be working on

This challenge is global in scope and does not focus on a specific project.