Ontology-Based Classification and Detection of the Smart Home Automation Rules Conflicts

Smart homes are the most adaptable and utilized application of the IoT. Smart homes enable end users to define and control the automation remotely by defining automation rules. The automation rules interaction results in the defined tasks and activities but also delivers unwanted interactions, which show adverse effects. Their interactions may consist of chained and covert rules interaction, which are hard to trace and identify. Attackers’ malicious apps installed in smart homes may leverage rules’ adverse effects to compromise smart home security and harm users, such as opening the window at night or when no one is home. To address such issues, we developed an ontology of smart home automation rules to identify such interactions. Prior works failed to provide the complete classification and could not identify all possible rule conflicts. In this work, we proposed the classification of automation rule interactions into five categories and recognized the potential rule conflicts. In addition, the ontology can examine rules interactions against the defined safety policies, providing an extra layer of security. The ontology will aid the designers in better understanding and developing a robust security mechanism at the design phase to resist rules interaction conflicts and their adverse effects. We examine this work on the thirty-five rules formed in the light of prior work test cases and validation. The results show that the proposed work is promising and can efficiently identify all the rule interaction conflicts in the smart home.