Article Preview
TopIntroduction
Internet of Things (IoT) has attracted the attention of major players in the industrial landscape and it is currently one of the most expected emerging technologies (Asghari et al., 2019).
Many studies (Asghari et al., 2019; Zdravković et al., 2016; Souri & Norouzi, 2019) suggest that the need for IoT platforms is mainly articulated around sector of consumer-centered enterprises, like smart homes and cities, services sector, industrial sector as well as transportation.
Obviously, all these platforms and technologies need an execution logic, acting as a command, coordinating and synchronizing different things together, to provide a plus value to the things composition. This execution logic has to be checked and validated before the implementation of the IoT platform because failures in such systems can lead to major damage. Hence, the use of a formal model is suited to model and verify the system before its effective deployment.
The modeling phase plays an important role in any project since it helps to understand how the system works. It provides a good way to deal with the complexity of systems and ensure their consistency.
On the other hand, the temporal aspect is an essential parameter in the interaction between the different things within the IoT platform. Therefore, the modeling paradigm should offer both temporal and logic modeling capabilities.
Like in service-oriented applications, IoT applications can be enhanced by composing existing atomic objects into composite ones to provide a benefit. IoT composition scenarios are applied to smart devices within an IoT context, to offer new applications in different fields.
Existing modeling languages for IoT applications development (Souri & Norouzi, 2019) remain general purpose; therefore, they cannot provide a detailed description of the application domain. Further customization of the modeling approach is needed with respect to domain-specific features. In this manner, domain-specific knowledge should be available to designers and could be reused across applications.
In order to address the lack of special-purpose modeling languages for IoT applications, the authors propose the Things-Net model. Things-Net is a temporal and hierarchical extension of Petri nets, it is based on the H-SMIL-Net model (Bouyakoub & Belkhir, 2008, 2011).
The Things-Net model is designed specifically to meet the needs of IoT-based systems. It allows an easy modeling of IoT time-dependent platforms, as well as a verification technique to ensure that the designed IoT solution fulfills the application's requirements. The model has three layers; each layer describes the system according to a different point of view, with a specific level of abstraction.
In order to illustrate the approach, the authors apply the proposed model to a prototype of a smart airport management system based on IoT, presented in a previous work (Bouyakoub et al., 2017).
The remainder of the paper is organized as follows: Section 2 presents a background about IoT-based systems and some related works. Section 3 is dedicated to the presentation of the Things-Net model. In section 4, authors apply the modeling approach to a case study: a smart Airport IoT based platform. Finally, Section 5 concludes this paper while giving some perspectives.