Intrusion detection by describing bus systems as stochastic processes

Umfeld:

The recent development explosion in sensor technology to support active safety, trends such as IoT, or autonomous driving make it essential to move away from traditional architectures consisting of networked ECUs or PLCs. The interconnectedness and powerfulness of centralized ECUs or central car servers plays in the hands of adversaries trying to gain access to the cars various communication and computing systems. Because of this, the detection of intruders in the car’s communication system can manifest as many different symptoms.

Aufgabe:

These symptoms can manifest themselves as small hiccups or large blackouts of the whole car. Detecting intruders can be done using stochastic methods and models. Detecting abnormal behavior by monitoring bus traffic is one strategy that caters to the predictiveness of real-world driving, while using trusted and proved methods. Detecting abnormal behavior by monitoring bus traffic is one strategy that caters to the predictiveness of real-world driving, while using trusted and proved methods.

Voraussetzungen:

• Very good knowledge of VHDL, Verilog (through HSC, DHL, or similar lectures) or alternatively in HLS languages such as SystemC, SystemVerilog or Chisel
• Good knowledge of describing system behavior of stochastic processes (e.g. Markov Chains, Petri Nets)
• Knowledge of communication systems (through CSP or similar lectures)