Author : David Reinoso
Publisher :
ISBN 13 :
Total Pages : 0 pages
Book Rating : 4.:/5 (138 download)
Book Synopsis Analysis of Influential Objects In Smart Home Platforms by : David Reinoso
Download or read book Analysis of Influential Objects In Smart Home Platforms written by David Reinoso and published by . This book was released on 2023 with total page 0 pages. Available in PDF, EPUB and Kindle. Book excerpt: Smart home platforms aim to provide convenience and peace-of-mind regarding home automation. With this quality of life improvement comes a necessity to ensure that smart homes remain secure and safe for the individuals residing within. Since home automation is deployed by a variety of methods (e.g., user-defined rules, local service applications, and integration between platforms) that may see different implementations between popular platforms, it is crucial that that a framework aiming to provide analysis for separate platforms be able to operate despite these differences. Furthermore, it is beneficial to acknowledge the importance of each entity (e.g., automation, device, abstract home object (AHO)) deployed in a smart home with respect to all other entities in the system. To this end, we propose the concept of influence with regards to entities in a smart home. Influence can be measured by the following three properties: (1) centrality, which corresponds to the overall connectivity of an entity in the network of all entities in a smart home, (2) transitivity, which highlights the entities most affected by other entities in the smart home, and (3) provenance, which identifies the entities that affect the most entities in the smart home. The use of `affect' refers to the ability for an entity to directly or indirectly access or control the state of another entity (e.g., a motion sensor is used that turns on a light `affects' that light, in the sense that an event originating from the motion sensor has effects on the light's state). The properties can be ascertained by evaluating a smart home's trigger-action routines (i.e., the encapsulation of automation in Smart Homes). We propose an approach to analyze the set of trigger-action routines extracted from a smart home defined based on the Samsung SmartThings platform. This analysis is composed of four stages in which automation in a smart home is extracted and formulated into a SELinux-inspired policy set, then the policy set is modelled as a flow graph which is analyzed for influence with respect to the three defined properties. We conduct an influence analysis using this approach on a smart home with 188 separate smartapps and 52 unique devices/AHOs defined on the Samsung SmartThings platform. The results of this analysis include three lists of top influential nodes with respect to these properties. These results are evaluated to yield key insight about entities in the smart home definition, and these insights can be projected as general security implications regarding high influence entities in smart homes. Namely, adversaries who gain control of entities with high centrality or provenance may compromise a major portion of the smart home, while adversaries who gain access to entities of high transitivity may learn private information about the smart home system. This analysis approach identifying such high influence entities provides a groundwork for addressing these and similar smart home security concerns as they become ever more relevant.