The concepts of the Internet of Things (IoT) and cyber-physical systems promise the ubiquitous availability of data. This enables applications to access data according to their rights and requirement profiles. Previous solutions implement self-contained applications whose data and control flow is statically configured. IoT devices transmit their data periodically or event-driven to a database that serves the requests of the applications. This abstracts the specific capabilities and performance spectrum of the IoT devices and therefore is not fully available for applications. With embedded systems, this is made even more difficult by the fact that a statically linked firmware is used, so that the chance to receive function and security updates is low.
In order to overcome the complexity and challenges above, DoRIoT consists of four working areas:
- Runtime environment
- Optimization of aggregation and disaggregation
- Security and safety
- Verification and validation
We transform the centralized architecture of existing SmartX environments into a dynamic architecture, transform static methods and tools into dynamic tools, and lay the foundations for emergent systems with organic computing methods. The use of organic computing allows the early detection of imminent failures or low service quality and the taking of appropriate countermeasures by (dis)aggregation of the affected services. Emergence is made possible by the creation of a uniform communication infrastructure that overcomes protocol boundaries with cross-protocol proxies and thus also incorporates existing infrastructure. The strategy of integrating existing products, interfaces and infrastructure is also the basis for the choice of the runtime environment. The illustration visualizes DoRIoT’s solution approaches in an exemplary SmartX environment.
The possibility of code migration creates new challenges for security and safety. Previous access control systems are always either centralized, and thus become a bottleneck, or they are not flexible enough to control the dynamics of the access rights of emergent systems. Both problems are addressed with a decentralized and dynamic rights management system. When services are migrated, it must be ensured that they continue to run correctly. Certifying algorithms decouple the calculation from the verification of the correctness of the results, whereby the verification can be carried out efficiently. Thus the verification can be efficiently implemented on trustworthy systems, while insecure systems can be securely used for complex calculations.
From an economic perspective, not only the reduced development costs resulting from the use of the developed tools and concepts are interesting. Two new value chains emerge: On the one hand, dynamic rights management allows premium services to be offered. On the other hand, the runtime environment allows extending existing hardware with new software functionality, which could, for example, be distributed via an “IoT App Store”.
In summary, DoRIoT provides the industry with tools and concepts with which the opportunities of networking in the environment of the IoT can be exploited and at the same time the risks can be controlled.