Inconsistency Detection-Based LOD in Smart Homes

Introduction

A smart house, through its numerous services, allows offering comfort to the inhabitants taking into consideration their preferences. It allows adapting its services according to their needs so as not to harm their health, such as keeping a humidity level for people suffering from asthma. It also offers the possibility to authorize or not a person to use an equipment or service, such as children or people with Alzheimer's. The inhabitants can even define scenarios to be performed, such as watching a movie or taking a bath. In the other hand, with the emergence of the IoT, the number of devices in a smart home continues to grow, offering users new services that will need to be adapted to their profiles (Schiaffino and Amandi, 2009). Indeed, each inhabitant has his own preferences and habits, which often creates conflicts. A conflict or inconsistency is the change from a coherent state to an incoherent state after the execution of a command. This inconsistency may concern the same equipment, so as one inhabitant opens the window and another wants to close it, or different equipment, for example, the first turns on the air conditioner while the second turns on the heating.

Several appliances can offer the same service; as to change the temperature of the house, we can use the air conditioner, the heater or the window. We must define the relation between the different appliances in the house. In addition, for better use of the devices, we need to know certain information, such as weather, air pollution, energy consumption, … etc. Even the health of the inhabitant is important in order to make the best decision. It is important to have a global view of the house and all the things that make it up, both the equipment and the person who live in it. For this, it seems obvious that the use of the Linked Open Data (LOD) (Bauer and Kaltenböck, 2016) is appropriate to determine the relation between these different elements. The LOD will allow us to exploit these relations in order to determine inconsistencies based on the antonyms of each element. Moreover, an antonym is defined using a thesaurus of antonyms. Indeed, the proposed solution, based on a semantic approach, consists in defining ontologies to contextualize each object in the house and the state of a room.

This paper is structured as follows: in the second section, we present related works to conflict resolution and management. In the section three, we explain the interest of a smart house based Software Defined Networks (SDN). The conflict management process is presented in the section four; we start by defining the user’s profile, scenario and services descriptions then the inconsistency detection process. The last part is about the conflict management solution. The simulation of our solution is presented in section five and the last section is reserved for the conclusion and the perspectives.