The term resilience is used in many scientific disciplines ranging from materials to psychology.
We are interested in the resilience of technical systems, focusing on cyber physical systems of systems (CPSoS). We are researching options to measure resilience, design and test resilient CPSoS. We define CPS(oS) as resilient, if they have the ability to react to specified and unspecified disturbances in a way that preserves their function and enables them to react quickly. This reaction includes the early detection, minimization, prediction or even avoidance of disturbances. The IHP has the following competencies in the field of resilient systems:
Competencies
-
Energy efficient implementations
Energy efficient implementations
An important strategy to reach resilience is redundency. In the application fields considered here such as e-health, automation networks, etc., embedded systems are generally used.
-
Tamper resistant crypto implementations
Tamper resistant crypto implementations
The protection goals confidentiality, data integrity and authenticity can be realized with the help of cryptographic algorithms. To achieve this, it is essential to keep the used keys secret.
-
Attack detection
Attack detection
The threat to IT systems from attacks is continuously increasing. The detection of attacks is very challenging especially for embedded systems as they have very limited resources.
-
Radhard RF and digital circuits for reliability critical applications
Radhard RF and digital circuits for reliability critical applications
Research activities for reliable digital and RF circuits for space application have long tradition. The major focus of this investigation is on radiation hard by design (RHBD) methodology for digital and RF circuits.
-
Adaptive fault tolerant multi-processors
Adaptive fault tolerant multi-processors
Processing is one of the critical operations in embedded systems with reliability relevant requirements. IHP is investigating single- and multi-processing systems that provide reliable operation for different applications, including space, automated driving, or industry automation.
-
Reliable computing acceleration for AI and DSPapplications
Reliable computing acceleration for AI and DSPapplications
Today AI applications are driving technology development. In order to fulfill performance and energy requirements, innovative solutions are needed.
Resilience is more than security and redundancy. Cognition makes the difference.
Security is THE prerequisite needed to ensure any feature of a certain system. Without security an attacker can take control of the system and in consequence all its features are subject to the intention of the attacker.
Redundancy is a well-known option to increase the reliability of almost any system. Without a certain level of redundancy a system will fail in the moment at which one of its essential components breaks. So at the least these components require redundancy. But the affordable redundancy is limited, e.g. 2 engines for certain types of airplanes. But what happens if the redundancy is lost, e.g. as in the case when an airplane experiences a double bird strike shortly after takeoff.
Coping with the Unexpected
What is going to help in an unexpected situation such as the double bird strike?
The answer is cognition, i.e. finding a solution to a formerly unexperienced problem. As we are working on CPSoS the challenge is not only to come up with a system that reacts properly to such a challenge, but to realize it with the very limited resources of embedded systems.
So metaphorically speaking, the problem we are trying to solve is squeezing Captain Sullenberger (the pilot who landed the airplane on the Hudson river) into millions of embedded systems.
