Caractéristiques
- Types de propriété intellectuelle : Brevet ; Logiciel
- Stade de développement : TRL3 - Preuve expérimentale de conception
-
Secteurs d'applications :
Industrie - Ingénierie - ProductionNumérique - Réseaux - Télécoms - SystèmesTransports
-
Domaines scientifiques :
SCIENCES DE L'INGÉNIEURSCIENCES ET TECHNOLOGIES DE L'INFORMATION ET DE LA COMMUNICATION
- Mots-clés : UAV ; Fleet ; Cybersecurity ; IoT wireless ; Networks ; Detecting intrusions
Description
Aircrafts as well as Unmanned Aerial Vehicles (UAV) are more and more organized as networks. Data exchange is becoming increasingly important with the need to perform complex functions: Traffic Management or Fleet Management (such as swarm applications). Maintaining the integrity of the aerial data network has become more and more crucial.
Wireless networks which topology can evolve randomly introduc[ing] additional security risks. To avoid undesirable consequences, it is necessary implement a security strategy.
The technology presented herein is a new design of intrusion detection and prevention system which is efficient against various security threats including unknown types of attacks.
Our technology is a methodology for detecting and preventing intrusions which can operate in real complex environments (such as air transport or UAV fleets). It is composed of three stages: detection, decision making and prevention.
• Intrusion detection is achieved through studying the signature of network. It is based on Wavelet Leader Multifractal Analysis and signature recognition
• Decisions are made to maximize the Quality of Service of the network under various types of attacks. Decisions are made upon the results from the signature recognition
• As patent application is pending, so far, no information can be given on the prevention system
This technology is efficient on wireless ad-hoc networks (i.e.: networks with evolutive topology) and enables the detection of 0-Day attacks.
Spécifications techniques
Key Technologies |
|
Advantages |
|
Avantages concurrentiels
• Detection of 0-Day attacks
• In depth traffic analysis
• New generation prevention system
• Efficient on wireless ad-hoc networks
• Adaptable to different types of networks
Champs d'application
• Air Traffic Management (ATM) / ADS-C
• UAV Traffic Management (UTM)
• Drones/Robots Swarm
• Autonomous Vehicles
• IoT and Industry 4.0