STewARD:SDN and blockchain-based Trust evaluation for Automated Risk management on IoT Devices

The Internet of Things (IoT) carries a big promise, to improve our lives through numerous connected devices that interact with one another. Unfortunately, some of these devices are of questionable security, becoming targets of choice for numerous exploits. This can result in compromising the home network, as well as large scale attacks such as the recent Mirai network of botnets that was used in a DDoS attack. The problem is that end users have a hard time assessing the risks induced by connected devices, and often lack the skills and time to administrate their home network. We answer the above challenge by proposing STewARD, a solution that allows users to easily request from their intelligent home network controller the creation of isolated software-defined network slices, to which they assign a required trust level using very simple risk assessment. We build a global trust assessment framework that computes a trust score for each class of devices, based on reported history stored in a blockchain. Network controllers can then leverage this information to connect to slices only those devices that meet the expected trust levels, and can contribute to the crowd-sourced reporting by monitoring the devices' behaviors compared to an expected baseline.