Adaptive systems of systems

In order to create a dependable, overall distributed system, the required flexibility must be taken into account when developing the system architecture.

Adaptive end-to-end architectures

The adaptive end-to-end architectures from Fraunhofer ESK can also be reliably used with safety-critical applications.

Cloud control

Outsourcing some functions to the cloud makes it possible to more easily implement life cycle management for machines.

Dependable distributed systems – adaptive systems of systems

The interaction of many individual systems within an overall connected and distributed system yields a wealth of challenges: new system participants such as devices, machines, applications and networks are constantly added or removed, or they fail. The quality of the connections fluctuates as a result, especially in dynamic environments. In order for these complex overall systems to function reliably despite the many uncertainties, flexibility must be taken into account when developing the system architecture. That means validating which functions can function where in the system. In this area, Fraunhofer ESK is working on the following approaches:

Adaptive end-to-end architectures for distributed systems

Applications must be continually dependable. That means they have to adapt while simultaneously adhering to all of the relevant specifications, such as time restrictions. The underlying technology must be also be used flexibly in order to react to uncertainties and to be able to guarantee that the application functions reliably. This is a significant factor particularly in environments such as connected driving or connected Industry 4.0 systems. With this in mind, we are working on methods for designing adaptive end-to-end architectures, which can also be reliably used for safety-critical applications, such as Quality of Service (QoS) monitoring and prediction for end-to-end wireless communications, among others.

Methods for Quality of Service (QoS) management

The foundation of a connected world is functioning communications system between the individual participants, such as in connected driving environments. To date, wireless communications quality has been difficult to manage. As a result, developers have to assume the worst case scenario when validating the functionality of the applications. We are working on methods that make the relevant QoS parameters predictable, thus allowing the system to proactively react on the basis of the prediction at the network or application level. We can thus create adaptive approaches for use by system and application developers based on the actual situation, instead of the worst case, even for safety-critical applications.

Privacy warning

With the click on the play button an external video from is loaded and started. Your data is possible transferred and stored to third party. Do not start the video if you disagree. Find more about the youtube privacy statement under the following link:

System-wide fail-operational connected function concepts

Within an overall system, the individual distributed systems have to constantly ensure the operation of their own safety-critical functions. At the same time, collectively the individual systems can enable high-quality functions. In this case it’s important that system reliability not be put at risk through such graceful upgrades. With this in mind, we are working on system-wide functions for fail operational concepts that make it possible to safely utilize external services from other sources such as edge, fog or cloud networks.

Runtime model and cloud control

With the advent of services-based architecture in Industry 4.0 environments, there will be an increasing opportunity to separate software-based functions and hardware. In order to easily implement life cycle management for connected and distributed systems and carry out over-the-air updates for instance, it makes sense to relocate individual functions to other networks such as the cloud. However, this requires making up-to-date information about the cyber physical systems available. This information ranges from long-term data for predictive maintenance, to process improvements and real-time data for cloud control. The latter has to be made available via runtime models in order to circumvent time-consuming maintenance activities. In this area Fraunhofer ESK is working on new methods for asset administration shells (AAS) and methods for the dependable management of mixed-critical applications from the cloud that enable centralized, distributed runtime administration.

Projects related to dependable distributed systems


2018 Drabek, Christian; Weiß, Gereon; Bauer, Bernhard:
Resumption of runtime verification monitors: Method, approach and application


Langmann, Reinhard; Stiller, Michael:
Cloud-based industrial control services: The next generation PLC

Oleinichenko, Oleg; Sevilmis, Yagmur; Roscher, Karsten; Jiru, Josef:
Time-controlled neighborhood-driven policy-based network selection algorithm for message dissemination in hybrid vehicular networks


Saad, Ahmad; Staehle, Barbara; Knorr, Rudi:
Predictive medium access control for industrial cognitive radio


Seydel, Dominique; Weiß, Gereon; Pöhn, Daniela; Wessel, Sascha; Wenninger, Franz:
Safety & security testing of cooperative automotive systems


Shekhada, Dhavalkumar; Stiller, Michael; Salvi, Aniket:
A comparison of current web protocols for usage in cloud based automation systems


Aktas, Ismet; Bentkus, Alexander; Bonanati, Florian; Dekorsy, Armin; Dombrowski, Christian; Doubrava, Michael; Golestani, Ali; Hofmann, Frank; Heidrich, Mike; Hiensch, Stefan; Kays, Rüdiger; Meyer, Michael; Müller, Andreas; Brink, Stephan ten; Petreska, Neda; Popovic, Milan; Ruchhaupt, Lutz; Saad, Ahmad; Schotten, Hans; Wöste, Christoph; Wolff, Ingo:
Funktechnologien für Industrie 4.0


Drabek, Christian; Weiß, Gereon:
DANA – Description and Analysis of Networked Applications


Drabek, Christian; Weiß, Gereon; Bauer, Bernhard:
Method for automatic resumption of runtime verification monitors


Fettweis, Gerhard P.; Franchi, Norman; Bittner, Frank; Dekorsy, Armin; Dillinger, Markus; Dyka, Zoya; Einsiedler, Hans J.; Fitzek, Frank; Frotzscher, Andreas; Glänzer, Martin; Hentschel, Tim; Hofmann, Frank; Hoffmann, Marco; Irmer, Ralf; Janßen, Uwe; Jiru, Josef; Jungnickel, Volker; Knorr, Rudi; Kraemer, Rolf; Kornbichler, Andreas; Kückelhaus, Markus; Langendörfer, Peter; Menges, Georg; Merz, Peter; Meyer, Michael; Mühleisen, Maciej; Müller, Andreas; Oswald, Erik; Ruchhaupt, Lutz; Redana, Simone; Reinartz, Michael; Richter, Klaus; Riedl, Johannes; Schotten, Hans; Schulz, Dirk; Schupke, Dominic; Thümmler, Christoph; Timm-Giel, Andreas; Wiebus, Christian; Willmann, Sarah; Zimmermann, Gerd:
Resiliente Netze mit Funkzugang


Franze, Juliane; Seydel, Dominique; Weiß, Gereon; Haspel, Ulrich:
Evaluation of traffic control systems as ITS infrastructure for automated driving


Petreska, Neda:
End-to-End Performance Analysis for Industrial IEEE 802.15.4e-based Networks
2017 Roscher, Karsten; Nitsche, Thomas; Knorr, Rudi:
Know thy neighbor - a data-driven approach to neighborhood estimation in VANETs


Petreska, Neda; Al-Zubaidy, Hussein; Staehle, Barbara; Knorr, Rudi; Gross, James:
Statistical Delay Bound for WirelessHART Networks
2016 Roscher, Karsten; Jiru, Josef; Knorr, Rudi:
Low-Delay Forwarding with Multiple Candidates for VANETs Using Multi-Criteria Decision Making
2016 Roscher, Karsten; Maierbacher, Gerhard:
Reliable message forwarding in VANETs for delay-sensitive applications


Saad, Ahmad; Mansour, Nour; Friedrich, Andreas; Youssef, Ziad; Dahlhaus, Dirk; Sharma, Mridula; Al Halaseh, Rana; Majeed, Erfan; Kohrt, Klaus D.; Bruck, Guido; Knorr, Rudi; Jung, Peter:
Cognitive Radio Prototype for Industrial Applications


Saad, Ahmad; Staehle, Barbara; Knorr, Rudi:
Spectrum Prediction Using Hidden Markov Models for Industrial Cognitive Radio
2015 Jiru, Josef; Mammu, Aboobeker Sidhik Koyamparambil; Roscher, Karsten:
Adaptive decision algorithms for data aggregation in VANETs with defined channel load limits


Petreska, Neda; Al-Zubaidy, Hussein; Knorr, Rudi; Gross, James:
On the recursive nature of end-to-end delay bound for heterogeneous wireless networks


Saad, Ahmad; Staehle, Barbara; Chen, Yun:
On interference detection using higher-order statistics


Petreska, Neda; Al-Zubaidy, Hussein; Gross, James:
Power minimization for industrial wireless networks under statistical delay constraints

Roscher, Karsten; Bittl, Sebastian; Gonzalez, Arturo; Myrtus, Matthias; Jiru, Josef:
ezCar2X. Rapid-Prototyping of Communication Technologies and Cooperative ITS Applications on Real Targets and Inside Simulation Environments


Saad, Ahmad; Staehle, Barbara; Auer, Alexander:
Real-time quality surveillance for industrial radio environments


Saad, Ahmad; Staehle, Barbara; Chen, Yun:
On the effectiveness of medium access with predictive collision avoidance


Petreska, Neda:
Towards new routing solutions for wireless industrial networks
2013 Steiner, Torsten; Roscher, Karsten; Jiru, Josef:
Cooperative glare reduction using V2V radio technology