On the Development of Systems-of-Systems based on the Internet of Things

Maia, Pedro; Cavalcante, Everton; Gomes, Porfírio; Batista, Thaís; Delicato, Flávia C.; Pires, Paulo F.

doi:10.1145/2642803.2642828

Cited by 10 publications

(12 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The National Academy of Science and Engineering emphasizes the evolution of self-contained embedded systems to Cyber-Physical Systems (CPS) into the Internet of Things (IoT) [69]. As a consequence, single system solutions become interconnected and finally, merge their pool of functionality into an overall System of Systems (SoS) [130]. A key characteristic for embedded systems is the interaction with the environment that can be a human or another system.…”

Section: Research Challenges and Goalsmentioning

confidence: 99%

“…First, because a SoS consists of several interconnected and diverse system types ranging from small embedded systems over cyber-physical systems to software intensive systems, it inherits such system characteristics as adaptiveness, collaborativity and flexibility. Second, a SoS is able to combine single capabilities by system interaction to offer new services or functionalities that cannot be realized by a single system alone [39,81,130,140]. Third, the complexity of the overall SoS raises challenges in describing, analyzing and understanding the emerged system behavior.…”

Section: Research Challenges and Goalsmentioning

confidence: 99%

“…Due to the causal connection, all changes in the runtime model will affect the running system itself, which allows an adaptive control of the software system on a much higher level of abstraction directly working with the corresponding runtime models [32,38]. Furthermore, because adaptive SoS are complex and inherent distributed [81,130] new challenges concerning the knowledge management arise. For example, if systems inside the adaptive SoS interact with each other to reach common goals, they must communicate with each other or share information for coordination purposes to agree on a common knowledge base for further task processing.…”

Section: Introductionmentioning

confidence: 99%

“…Thus, the modeling language must have a clear semantic for the runtime models itself, its access by the adaptation logic as well as between different systems inside the SoS. Even worse, an adaptive SoS is inherent distributed with emergent behavior [39,81,130,140]. Therefore, contained systems must collaborate with each other to reach SoS goals by simultaneously optimize their own behavior according to local goals.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Modeling Collaborations in Adaptive Systems of Systems

Wätzoldt

Giese

2015

Proceedings of the 2015 European Conference on Software Architecture Workshops

View full text Add to dashboard Cite

Recently, due to an increasing demand on functionality and flexibility, beforehand isolated systems have become interconnected to gain powerful adaptive System of Systems (SoS) solutions with an overall robust, flexible and emergent behavior. The adaptive SoS comprises a variety of different system types ranging from small embedded to adaptive cyber-physical systems. On the one hand, each system is independent, follows a local strategy and optimizes its behavior to reach its goals. On the other hand, systems must cooperate with each other to enrich the overall functionality to jointly perform on the SoS level reaching global goals, which cannot be satisfied by one system alone. Due to difficulties of local and global behavior optimizations conflicts may arise between systems that have to be solved by the adaptive SoS.This thesis proposes a modeling language that facilitates the description of an adaptive SoS by considering the adaptation capabilities in form of feedback loops as first class entities. Moreover, this thesis adopts the Models@runtime approach to integrate the available knowledge in the systems as runtime models into the modeled adaptation logic. Furthermore, the modeling language focuses on the description of system interactions within the adaptive SoS to reason about individual system functionality and how it emerges via collaborations to an overall joint SoS behavior. Therefore, the modeling language approach enables the specification of local adaptive system behavior, the integration of knowledge in form of runtime models and the joint interactions via collaboration to place the available adaptive behavior in an overall layered, adaptive SoS architecture.Beside the modeling language, this thesis proposes analysis rules to investigate the modeled adaptive SoS, which enables the detection of architectural patterns as well as design flaws and pinpoints to possible system threats. Moreover, a simulation framework is presented, which allows the direct execution of the modeled SoS architecture. Therefore, the analysis rules and the simulation framework can be used to verify the interplay between systems as well as the modeled adaptation effects within the SoS. This thesis realizes the proposed concepts of the modeling language by mapping them to a state of the art standard from the automotive domain and thus, showing their applicability to actual systems. Finally, the modeling language approach is evaluated by remodeling up to date research scenarios from different domains, which demonstrates that the modeling language concepts are powerful enough to cope with a broad range of existing research problems.-I - Zusammenfassung

show abstract

Section: Research Challenges and Goalsmentioning

confidence: 99%

Section: Research Challenges and Goalsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Modeling Collaborations in Adaptive Systems of Systems

Wätzoldt

Giese

2015

Proceedings of the 2015 European Conference on Software Architecture Workshops

View full text Add to dashboard Cite

show abstract

“…However, interoperability and emergence are such intrinsic properties of IoT applications that the classic notion of a system is inadequate for dealing with system complexity, and tackling security problems. In recent years, there has been some interest in framing IoT systems as Systems-ofsystems [2]. A System of Systems (SoS) is an integration of independent and operable constituent systems, networked together to achieve a higher goal [3].…”

Section: Introductionmentioning

confidence: 99%

From Requirements to Operation: Components for Risk Assessment in a Pervasive System of Systems

Ki-Aries

Doğan

Faily

et al. 2017

2017 IEEE 25th International Requirements Engineering Conference Workshops (REW)

View full text Add to dashboard Cite

Abstract-Framing Internet of Things (IoT) applications as a System of Systems (SoS) can help us make sense of complexity associated with interoperability and emergence. However, assessing the risk of SoSs is a challenge due to the independence of component systems, and their differing degrees of control and emergence. This paper presents three components for SoS risk assessment that integrate with existing risk assessment approaches: Human System Integration (HSI), Interoperability identification and analysis, and Emergent behaviour evaluation and control measures. We demonstrate the application of these components by assessing a pervasive SoS: a SmartPowerchair.

show abstract