What does trust mean? For the purposes of this book, perhaps the simplest definition would be that trusted computing refers to a computer system for which an entity has some level of assurance that (part or all of) the computer system is behaving as expected. The entity may be any one of various things, including the human user of the PC or a program running on a remote machine. The degree of coverage of this assurance, that is, whether it covers all aspects of the system or just some part, and the nature of the entity to which assurance is provided, vary depending on the system and the environment within which it is used.

In this chapter, the state of the art in trusted computing technology has been examined. Various initiatives, projects and specification sets have been explored. It is clear that, despite the negative criticism often associated with this particular area of computing, trusted computing technologies offer a wide range of functionality which may be leveraged to improve computer security. It must also be noted, however, that this chapter reflects these technologies as they are currently documented. This area is the subject of much current research and development, and the specifications, functionality, architectures, mechanisms and implementations associated with trusted computing technologies are evolving and changing very rapidly.

This chapter describes the Direct Anonymous Attestation (DAA) scheme in the con text of its development and use within the TCG specification. This scheme was originated by Brickell, Camemsch and Chen. It was adopted by the Trusted Computing Group as the method for remote anonymous authen tication of a Trusted Platform Module (TPM). In this chapter, we first talk about how the work of designing DAA started from a conflict between the need for security and the need for privacy in TPM authentication. In the historical context, we acknowledge various schemes which have been developed in response to this TCG problem, and which eventually led to the DAA scheme. In the context of a solution to the problem. we focus on explaining how the DAA scheme proposed by Brickell, Camemsch and Chen works. In the context of further consideration, we discuss some options for the reissue of DAA keys and some possible improvements to increase the performance of the DAA.

One of the driving forces behind recent developments in mobile communications systems is the potential for such systems to deliver more complex content to con sumers. Current 3G systems are capable of delivering multimedia clips to subscribers' mobile telephones. The next generation of communications systems are expected to develop this service, and collaborate with broadcast systems to provide wireless access to video content from a wide range of mobile devices. For a service like this to achieve its full commercial potential, the owners of the content will require assurance that their material is not illegally accessed. Current broadcast systems accomplish this by using conditional access systems to ensure that only bona fide subscribers have access to the content.

In this chapter, a novel way of delegating a TTP service from a central TTP to a secure execution environment (SEE) running on a user's machine is proposed. It also shows significant advantages for some types of TTP service and, in particular, it may be a useful technique for solving some of the problems traditionally associated with certificate authorities (CAs).

One can expect a number of new applications that support the management of large networks, for example, to automatically verify that all machines have a proper patch level or only contain company authorised software, to maintain control over signature and encryption keys, etc. As trusted computing (TC) technology exists on the lowest hardware level of the computer, it has to be further supported by higher-level functionality such as the operating system before mainstream applications can fully profit from the concept. Thus, outside of such applications and those that use cryptography anyway, little use will be made of TC in the near future. The candidate application that always most commonly appears in the press and various Internet discussions is the use of TC for digital rights management (DRM), i.e. the option to attach a policy to a piece of data and remotely enforce this policy. While research in this area is becoming rather active, the author is not aware of any attempts to use Trusted Computing Group (TCG) technology to this end on a personal computer (PC) platform. It is debatable if the TC concepts can significantly assist a DRM system on a PC platform without massive changes to the underlying operating system. Without some significant innovation, a DRM system using the Trusted Platform Module will probably not be much more secure than one without.