Bachelor thesis: Software defined network infrastructure (gn)
Die Zukunft.

Bei Interesse können Sie uns Ihre Bewerbung übermitteln.

Bachelor thesis: Software defined network infrastructure (gn)
Standort: Ottobrunn bei München
Beschreibung der Stelle: IABG Innovation Centre is a development incubator for its IABG's portfolio, which, among others, includes data analysis, automatisation, and optimisation in the mobility and security domain. Natural Language Processing (NLP) has become one of our key competences and we would like to expand on this with you by applying it to one of our main projects in the domain of cyber security.

As part of the ongoing IMMUNE research project, IABG develops a risk assessment tool for industrial networks. Sensor information is interpreted to assess the current state of the network in terms of an exploitation of system vulnerabilities to derive guidance for risk minimisation. Vulnerabilities hereby include software/hardware flaws or configuration errors, which an attacker can employ to gain access. Thus, an attacker may propagate through the system via lateral movement and eventually reach a targeted machine. The sensors mainly perceive information coming from network based intrusion detection systems such as traffic load, connection geometry, and network topology. This state information is used to infer countermeasures defined on the action space upon the software-defined network configurations and the virtualized network function deployments.
  • Get familiar with software defined networks and Bro (Publish subscribe service).
  • Set up a virtual environment to deploy a simplified industrial network.
  • Model the action and state space of the network.
  • Derive meaningful topics for Bro.
  • Broadcast the state and action space using Bro.
  • Good programming skills (essential)
  • Experience with natural language processing (desirable)
  • Good communication skills in English
  • Ability to work both independently and collaboratively
Ziele: The goal of this Bachelor thesis is the simulation of a software defined network using Openflow and Mininet in a virtual environment and to provide an interface with tools to manipulate the state space, among others, the topology and status of the network and components of the network. Further, the action space, e.g. 'migrate component x to subnetwork y', 'detach component x', 'cut connection between component x and z', 'reinstall software s on component x', needs to be derived and implemented.

The result will be implemented into our IMMUNE Risk Assessment tool, which will be used within our company for consulting purposes.

This thesis lies the foundation of a virtual environment in which adversarial reinforcement learning will be applied to train the system in order to defend itself from attacks.

Bei Fragen hilft Ihnen unser Recruitment-Team Tel. 089 6088-2070 gerne weiter.

INFONIQA Informationstechnik GmbH   Textbaustein "FOOTER_RIGHT"