Towards an extensible privacy analysis framework for Smart Homes

Abstract

The IoT ecosystem is an intricate and complex network of stake- holders that includes platforms, developers, ad networks and cloud providers. However, the ability of smart home platforms and devices to interact and exchange data, together with the data-driven busi- ness models adopted by most IoT stakeholders open the ground for unknown and unexpected privacy risks. Existing black-box testing approaches to audit IoT platforms cannot identify data dissemina- tion through side- and covert-channels, and for this reason they are not well suited for rich execution environments where a wide range of devices and applications can co-operate using multiple network protocols and interfaces. This poster proposes ImposTer, a cost-effective and extensible privacy framework for exhaustively testing the IoT ecosystem. Our framework is able to capture, model and emulate horizontal interactions that occur across the different devices in a consumer household.