Multi-Layer Agent Based Architecture for Internet of Things Systems

Multi-Layer Agent Based Architecture for Internet of Things Systems

Kouah Sofia (RELA(CS)2 Laboratory, University of Larbi Ben M'Hidi, Oum El Bouaghi, Algeria) and Kitouni Ilham (MISC Laboratory, MISC Laboratory, Constantine 2-Abdelhamid Mehri University, Algeria)
Copyright: © 2018 |Pages: 21
DOI: 10.4018/JITR.2018100103

Abstract

Nowadays, the Internet of things (IoT) is becoming a promising technology which revolutionizes and simplifies our daily life style. It allows interaction and cooperation between a large variety of pervasive objects over wireless and wired connections, in order to achieve specific goals. Moreover, it provides a concise integration of physical world into computer systems through network infrastructure. This paper provides an agent-based architecture for developing IoT systems. The proposed architecture is multi-layer and generic. It encompasses four layers: Physical Component Management, Local Management -Coordination, Global Management-Coordination and Specialized Operative Management Layers. The first one can be seen as a smart layer that ensures connection and communication between things and the system. The second one constitutes the intelligent core of the system which acts locally to ensure coordination and further internal functioning. The third layer ensures coordination between the local system and the externals ones. The last layer supports additional behaviors which are domain dependent. The architecture is illustrated by an IoT system diagnosis.
Article Preview
Top

1. Introduction

Internet of Things is an attractive advanced research area. It involves peoples, everyday devices, software, hardware, sensors, actuators and pervasive objects to build an efficient system by means of network; as a communication medium. Such connected objects (i.e. things) are able to communicate, interact and exchange relevant data with one another by using embedded sensors (Terry, 2016). These data can be analyzed and treated differently with respect to application domain and system functionalities (i.e. main goal).

These emergent competences enable real time continuous access and transmission of information about a system or device in both natural and physical systems as well as in personnel and industrial environments (Welbourne, 2009). A few examples include: individual, private and surrounding environment (wearable devices and smart phones), smart home (home security devices, appliances), power grid (sensor and actuator in the power distribution systems), water systems (distribution grids, asset management and preventive maintenance; ambient environments), traffic and transportation (sensors on vehicles, sensors in roads, traffic lights and congestion control devices), structural health monitoring and health care (structures, vehicles, health devices), oil and gas.

The IoT systems go beyond the connectivity supported by a network of huge number of communicating objects via the Internet. They also link the space, time, events, processes, resources, etc. Currently, these characteristics are implanted in often dispersed digital models or systems, which are designed for different purposes. The encapsulation of these isolated attributes into a single virtual system is a major challenge. It may offer the possibility of transforming the practice of management of daily lives in new digital planet.

Pointing out that within IoT domain, the system is widely open, dynamic and needs large amount of communications, interactions and data exchanges that should be achieved in a coherent manner. It is a system consisting of heterogeneous physical and virtual connected components, using different languages, platforms and intelligent policies (IoT, 2015). As Wegner argued in (Wegner, 1997), interaction allows limitless possibilities for objects to control and access their environment and compensates for a lack of self-sufficiency. Connected to the internet, things acquire intelligence since they tap into an exceedingly rich environment. To exploit this potential, a precise and well-defined description as well as modeling of a massive and heterogeneous collection of interfaces, protocols and components, are required.

According to these properties, the domain of designing IoT systems is becoming increasingly attractive; since several IoT functionalities have to be modeled. Therefore, our paper focuses on the development of IoT systems.

Generally, the development process of any computer system intends to define the architecture, components, interfaces and modules that satisfy system requirements. In other words, it considers the main significant aspects and takes into account the required specifications in order to produce proficient systems; which are based on strong technical, analytical and development designers’ skills.

Although, the domain of developing IoT system is innovative for the imminent future of computing and communication, the development of effective and well-performing IoT systems still faces many challenging issues. Few approaches have been proposed in the literature, such as (Katasonov, Kaykova, Khriyenko, Nikitin & Terziyan, 2008; Kazanavicius, Kazanavicius & Ostaseviciute, 2009; Chen, Tseng, Lian, 2010; Fortino, Guerrieri & Russo, 2012; Ayala, Amor & Fuentes, 2012). These approaches are ad-hoc and do not support almost IoT systems functionalities and properties. In particular, they are application domain dependent and do not take into account the intelligent feature which is quite important to manage and control properly the objects’ traffic flow and improve decision making. Therefore, new models, paradigms and tools are required.

Complete Article List

Search this Journal:
Reset
Open Access Articles
Volume 14: 4 Issues (2021): Forthcoming, Available for Pre-Order
Volume 13: 4 Issues (2020)
Volume 12: 4 Issues (2019)
Volume 11: 4 Issues (2018)
Volume 10: 4 Issues (2017)
Volume 9: 4 Issues (2016)
Volume 8: 4 Issues (2015)
Volume 7: 4 Issues (2014)
Volume 6: 4 Issues (2013)
Volume 5: 4 Issues (2012)
Volume 4: 4 Issues (2011)
Volume 3: 4 Issues (2010)
Volume 2: 4 Issues (2009)
Volume 1: 4 Issues (2008)
View Complete Journal Contents Listing