Certificate Management Using Distributed Trusted Third Parties

Trust is a key component in any ubiquitous computing system. Users have to trust the devices to be secure, devices have to authenticate the users in order to trust their inputs and devices have to trust each others’ identity and authorisation. A central question in dealing with trust is how to distribute copies of a user’s public key in such a way that other users can verify that it does, indeed, belong to the user that claims ownership. Traditional answers to this question have involved using a trusted Certificate Authority (CA) to generate and distribute digitally signed certificates that bind a user’s name to his public key (and any other data that may be required). However, the centralised CA model is particularly unsuited to the rapidly changing, ad hoc network topologies that are associated with ubiquitous computing environments. Despite these difficulties, several authors have attempted to produce models and security analyses for CA systems in ubiquitous computing environments. The problems of implementing a “personal PKI” are discussed by Mitchell and Schaffelhofer [3] and can be applied to personal distributed environments [1]. Solutions for various network models have been proposed by Zhou and Haas [8], Luo and Lu [2], Varadharajan et al [7] and Zouridaki et al [9]. Many of which rely on the distribution or sharing of CA functionality between various devices within the network in an ad hoc fashion. We propose a simple solution for a distributed CA in a ubiquitous computing environment based on emerging secure execution environment technology [5]. For our purposes, we assume the existence of a highly resilient secure execution environment (SEE) with a hardware protection mechanism. The requirement that the SEE be hardware based is necessary to provide the additional protection that must be afforded to cryptographic data and differentiates such environments from the less-well protected software based environments such as the Java sandbox. Hardware based SEEs are gradually becoming available to low end users in the form of the trusted computing platform standardised by the Trusted Computing Group [6] and Microsoft’s NGSCB [4]. We assume that an SEE has three basic properties: (1) it can demonstrate to a third party that is has been initialised securely; (2) it can, and is able to demonstrate to a third party that it can, download applications in a secure fashion; (3) it can, and is able to demonstrate to a third party that it can, execute applications supplied to it by a third party. Essentially, an SEE provides an environment in which a third party can install a “black box” application on a remote machine without compromising the security of the application or the machine. Our solution to the problem of running a CA in a ubiquitous computing environment is to allow every user in that environment to download a “CA applet” — a self-contained application that will run on the user’s SEE and will issue certificates for that user’s public keys (and, potentially, other users that have been authorised by a pre-determined policy). Furthermore, that applet may, optionally, take the role of the directory service and make these certificates available to other network users. Hence, these CA applets may be placed anywhere within a network’s topology, as required by either the user or by some sort of controlling entity. The actual mechanics of installing a CA applet are very simple: a user requests a CA applet from a central trusted third party (TTP) which checks the user’s identity and that the user has a suitably secure SEE; the TTP then generates a new signature key pair for the new CA applet and issues a certificate for the applet’s public key; then the central TTP securely downloads the CA applet onto the user’s SEE where it is executed. At this point the user no longer needs to