Moving target defense for Internet of Things using context aware code partitioning and code diversification
Kaleel Mahmood, Devu Manikantan Shila · 2016
It is estimated that by 2020 there will be 50 billion connected IoT devices [1]. While the IoT has the potential to provide huge improvements in our daily life, it also introduces new security vulnerabilities. Many IoT devices have been shown to be easily exploitable. This is mainly due to the static footprint of the devices. Moving Target Defenses (MTDs) have been recently proposed as an approach to dynamically change the properties or configuration of a system, which greatly reduces an attacker's ability to successfully exploit that system. Directly applying MTD strategies may be infeasible for resource constrained IoT devices. We propose a new compositional MTD defense for IoT devices based on context aware code partitioning and code diversification. In this approach the devices will have minimal trusted code residing on the device. When context dictates, these devices request code and download it from a secure source (e.g., Cloud). The downloaded code is then loaded and run. After use, the code on the device is erased. In addition to code partitioning, we apply code diversification to also mitigate security attacks. Our final proposed system is a resource efficient high security framework for IoT devices.