Scalable Trust Assessment and Remediation of Wireless Devices
In large scale deployments of partly autonomously communicating and connecting network elements, such as the Internet of Things and machine-to-machine devices, trust issues have new qualities. Concurrently, end-user devices are technically open platforms, and also pose security threats on a large scale to users and networks. Thus, fault detection and remediation methods become costly. It is a key challenge to balance the requirements of scalability and cost-effectiveness with desired fine-grained checks and remote remediation. Current technologies, such as Trusted Computing Group’s Trusted Network Connect and Open Mobile Alliance’s Device Management Standards, may not be an ideal fit to the requirements. Extensions like property-based attestation (PBA) are promising, but may require special infrastructure and/or further standardization. We consider an architecture for Platform Validation and Management (PVM) in which designated network entities protect the access network by remotely validating devices before they are allowed to authenticate and gain access. We propose methods to diagnose devices with a granularity which allows also attachment even with partial functionality and methods to remediate faulty devices remotely, i.e., bring them back into a known good state. This approach requires some separation of tasks between network PVM entities and trusted functionalities on devices [1, 2]. Our generic and efficient approach to PVM, rests on three key ingredients:
First, A trusted platform architecture allowing separation of the system in a Secure Execution Environment (SEE) and a Normal Execution Environment (NEE). The system is capable of performing a secure start-up (bootstrap) process anchored in a Root of Trust (RoT) and building a chain of trust from the RoT to SEE to NEE, verifying start-up, particularly started components against Trusted Reference Values (TRVs). This is an abstraction of the Trusted Computing Group’s MPWG (Mobile Phone WG) Platform Architecture, which may be mapped to many different, concrete architectures.
- 3.Schmidt, A.U., Leicher, A., Shah, Y., Cha, I.: Tree-formed Verification Data for Trusted Platforms, http://arxiv.org/abs/1007.0642v3